input
stringlengths 4.25k
4.38k
| output
sequencelengths 1
1
| id
stringlengths 40
40
|
|---|---|---|
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: bookcase.
Question: We tried to get the bookcase through the door, but the _ was not wide enough.
Answer: door.
Output:
| [
"We tried to get the bookcase through the door, but the _ was not narrow enough."
] | task034-adb893eab871489eb87ba1dab7f3e936 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: China.
Question: For the dinner party, Alex wanted to use their china plates not the regular ones. His wife said no because the _ plates had to be handwashed.
Answer: china.
Output:
| [
"For the dinner party, Alex wanted to use their china plates instead not regular ones. His wife said no because the _ plates could be dishwashed."
] | task034-bd2f8137e9bd4a8692c64af46b115c33 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: mountain.
Question: Randy wanted to vacation in either the mountains or the beach. He chose the _ since it was cold.
Answer: mountains.
Output:
| [
"Randy wanted to vacation in either the mountains or the beach. He chose the _ since it was hot."
] | task034-862186adf50b42b59f2ba2f596baa4cc |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: skin.
Question: The skin of a human is smoother than the hide of an animal because the _ is always moisturized.
Answer: skin.
Output:
| [
"The skin of a human is smoother than the hide of an animal because the _ is always dirty."
] | task034-93abe7276d02465aa69d5eadc6982c83 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: drink.
Question: She ordered ginger ale to drink instead of lemonade because the _ tasted bad to her.
Answer: lemonade.
Output:
| [
"She ordered ginger ale to drink instead of lemonade because the _ tasted good to her."
] | task034-4c0b57cc881a472ea25be2ffcf863b0d |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: box.
Question: The ball had a lot of room to move around when placed in the box. The _ is small.
Answer: ball.
Output:
| [
"The ball had a lot of room to move around when placed in the box. The _ is big."
] | task034-c4507f5c14aa40cfa7a30b345bb19d5c |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: frog.
Question: The frog likes warmth and prefers to sit on the laptop versus the desk. The _ must be warmer.
Answer: laptop.
Output:
| [
"The frog likes warmth and prefers to sit on the laptop versus the desk. The _ must be cooler."
] | task034-948744f2b89c436c85e518f2e3601e89 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: rat.
Question: The rat took fabric from the quilt and put it in the nest, so now the _ has less material.
Answer: quilt.
Output:
| [
"The rat took fabric from the quilt and put it in the nest, so now the _ has more material."
] | task034-66b5191e290b46bbaeff98a71becfabd |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: psychologist.
Question: The psychologist had a desk and a couch, he sat at the _ during the session.
Answer: desk.
Output:
| [
"The psychologist had a desk and a couch, his patient sat at the _ during the session."
] | task034-87d1259e0a1d4acb83a67f1755a56bfd |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: money.
Question: Josh wanted to stop spending all of his money and make a budget, so he put the _ into a bank.
Answer: money.
Output:
| [
"Josh wanted to stop spending all of his money and make a budget, so he put the _ into a program."
] | task034-8a8b87c39d7f460c849044b1738e356a |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: shawl.
Question: The crocheted shawl would need a chain added onto it because the _ was unstructured.
Answer: shawl.
Output:
| [
"The crocheted shawl would need a chain added onto it because the _ was structured."
] | task034-93e845c22842491bb4b7228ed09bfc5b |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: house.
Question: Dan is looking to buy a house in countryside and not in City because the _ is more dense.
Answer: city.
Output:
| [
"Dan is looking to buy a house in countryside and not in City because the _ is more sparse."
] | task034-7be1a55e2aa74dfa81b3dea2022bb0fe |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: Fever.
Question: The ice placed on James head could not bring down his fever because the _ is too small.
Answer: ice.
Output:
| [
"The ice placed on James head could not bring down his fever because the _ is too high."
] | task034-8ed3481dda4544928b281305d2ac7f01 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: origami.
Question: I wanted to learn origami, but I couldn't find the paper or a book that worked for me. Eventually I bought the _ and read it.
Answer: book.
Output:
| [
"I wanted to learn origami, but I couldn't find the paper or a book that worked for me. Eventually I bought some of the _ and tried it."
] | task034-24e2101be0504b8ba4442b011e7bc6a7 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: math.
Question: Math was a harder subject for Susan than English was. That's because _ was confusing to her.
Answer: math.
Output:
| [
"Math was a harder subject for Susan than English was. That's because _ was clear to her."
] | task034-3f494884d7514baaa842533e1de0e6b3 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: phone.
Question: At the mall, Angela gave the phone to her mom, who put it in her purse because the _ was too heavy.
Answer: phone.
Output:
| [
"At the mall, Angela gave the phone to her mom, who put it in her purse because the _ would relieve the weight."
] | task034-d7f8bb7c86bc49b19875cd3cf6976b36 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: hug.
Question: Roger gave his mother a hug after he took a shower instead of after he played in the sandbox because the _ was dirty.
Answer: sandbox.
Output:
| [
"Roger gave his mother a hug after he took a shower instead of after he played in the sandbox because the _ made him clean."
] | task034-4ccd38a8581b413eb4d34aaba78e96b2 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: posture.
Question: Mary wanted to improve their posture at work, so they turned down the computer brightness and adjusted their chair because the _ was too bright.
Answer: computer.
Output:
| [
"Mary wanted to improve their posture at work, so they turned down the computer brightness and adjusted their chair because the _ was too low."
] | task034-2d5174b2d58a45e9aaee8545a37033ab |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: knife.
Question: George opted for both of them to use a knife instead of a gun in the duel because the _ could mortally injure them.
Answer: gun.
Output:
| [
"George opted for both of them to use a knife instead of a gun in the duel because the _ could partially injure them."
] | task034-e020f89ad28b47f487e051532e2c4795 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: marbles.
Question: He thought he lost his marbles in the dream world, but the _ were in his pocket the entire time.
Answer: marbles.
Output:
| [
"He thought he lost his marbles in the dream world, but the _ played tricks on him the entire time."
] | task034-b498753efbfb415bbd71a681d3bb04c5 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: afford.
Question: I could not afford to be late to the train or I would be late to a meeting, fortunately the _ was running late so I made it on time.
Answer: meeting.
Output:
| [
"I could not afford to be late to the train or I would be late to a meeting, fortunately the _ was running early so I made it on time."
] | task034-f804ef5999c94bdfbf6fbb86804273e9 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: fever.
Question: David had a high fever and planned to put medicine in his soup, because the _ tasted awful.
Answer: medicine.
Output:
| [
"David had a high fever and planned to put medicine in his soup, because the _ tasted good."
] | task034-d133f180c50341618d39c791d1d8fbcb |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: pelts.
Question: Amy used a whipstitch to sew the pelts together with cotton thread instead of nylon thread because the _ is more manageable.
Answer: cotton thread.
Output:
| [
"Amy used a whipstitch to sew the pelts together with cotton thread instead of nylon thread because the _ is less manageable."
] | task034-33de7b5f7f374c89b9a617c80e0eaaef |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: Life.
Question: Thinking about my life I have to choose between going to school or working. The _ approach would give me more money.
Answer: working.
Output:
| [
"Thinking about my life I have to choose between going to school or working. The _ approach would make me more intelligent."
] | task034-768ce4e2d1d040c0a474c4b5e770c69d |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: pours.
Question: The man pours the water into the bottle instead of the glass because the _ is cracked.
Answer: glass.
Output:
| [
"The man pours the water into the bottle instead of the glass because the _ is unbroken."
] | task034-ee6fba339ea74e0585864c150cb52ff1 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: materials.
Question: Ava wanted to pursue a consulting career, so she paid the fees for training materials, hoping to earn sales commissions. The _ turned out to be very large.
Answer: fees.
Output:
| [
"Ava wanted to pursue a consulting career, so she paid the fees for training materials, hoping to earn sales commissions. The _ turned out to be very small."
] | task034-0d8d2a8d55f445ba9db3819e60fb8d3b |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: Groom.
Question: To groom his facial hair, Adam preferred to use a hand razor over an electric razor because the _ was too rough on his skin.
Answer: electric razor.
Output:
| [
"To groom his facial hair, Adam preferred to use a hand razor over an electric razor because the _ was gentle on his skin."
] | task034-73be2b569df94a10a4ee8ab546baf592 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: candy.
Question: The candy broke apart easily with just a blow on it. The _ is too weak.
Answer: candy.
Output:
| [
"The candy broke apart easily with just a blow on it. The _ is too strong."
] | task034-f17d7118d209418a9653ca8b52973ad9 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: lumber.
Question: Mike tried to carry a stack of lumber to his truck but he could not because the _ was too wide.
Answer: stack.
Output:
| [
"Mike tried to carry a stack of lumber to his truck but he could not because the _ was too narrow."
] | task034-0fe11b4cbd124df3817bb6fb64fb1296 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: book.
Question: A book was loaned out for a long homework assignment but the _ was too basic.
Answer: book.
Output:
| [
"A book was loaned out for a long homework assignment but the _ was too complicated."
] | task034-265a921c47754ca6828395b7ebd63dd9 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: blade.
Question: The razor blade got broken when sharpening the boy's pencil because the _ is weak.
Answer: blade.
Output:
| [
"The razor blade got broken when sharpening the boy's pencil because the _ is hard."
] | task034-97b9787137b642e69774521c833fe28f |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: flowers.
Question: The stems of the fresh cut flowers would not fit in the vase because the _ is too small.
Answer: vase.
Output:
| [
"The stems of the fresh cut flowers would not fit in the vase because the _ are too long."
] | task034-379a7b509e6a4ffeaa7049b1a4aef2cc |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: guitar.
Question: I played the guitar better than I played the piano because I practiced more on the _ .
Answer: guitar.
Output:
| [
"I played the guitar worse than I played the piano because I practice more on the _ ."
] | task034-f05096d19477434b981051fbb4c96b1d |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: landlord.
Question: My landlord made more money renting the houses than the apartments because the _ were smaller.
Answer: apartments.
Output:
| [
"My landlord made more money renting the houses than the apartments because the _ were bigger."
] | task034-77591aa347bd4ff886172afdbd6bc03d |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: flower.
Question: Lucy grew roses and daffodils in her flower garden, but liked the _ best because red was her favorite color.
Answer: roses.
Output:
| [
"Lucy grew roses and daffodils in her flower garden, but liked the _ best because yellowyell her favorite color."
] | task034-7d8a5395597248428b405aae8822cc81 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: trap.
Question: The used a box instead of the trap to catch the mouse because the _ was more humane.
Answer: box.
Output:
| [
"The used a box instead of the trap to catch the mouse because the _ was more inhumane."
] | task034-0c2f1af2d797434585ea4077f0dda751 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: cooked.
Question: The steak cooked faster than the roast did because the _ was a thinner piece of meat.
Answer: steak.
Output:
| [
"The steak cooked faster than the roast did because the _ was a thicker piece of meat."
] | task034-211398b32c1444ba97398b375f01ca0a |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: canvas.
Question: I thought the canvas was big enough to fit the patio, but it didn't because the _ was too small.
Answer: canvas.
Output:
| [
"I thought the canvas was big enough to fit the patio, but it didn't because the _ was too big."
] | task034-434bcf9ba9ad4d799b192f6146ea006e |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: closet.
Question: The clothes would not fit in the closets after we bought them since the _ were too spacious.
Answer: clothes.
Output:
| [
"The clothes would not fit in the closets after we bought them since the _ were too cramped."
] | task034-a3a4ec84cd7241138f848b29d0218fe8 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: husband.
Question: My husband had to choose between a watch and a pen as his gift, he chose the _ because he liked to have many good ones.
Answer: pen.
Output:
| [
"My husband had to choose between a watch and a pen as his gift, he chose the _ because he liked to have a few good ones."
] | task034-f2cd230223854da39a4645085488d031 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: computer.
Question: The computer costed more to fix than the phone because the _ was much newer.
Answer: computer.
Output:
| [
"The computer costed more to fix than the phone because the _ was much older."
] | task034-49ab12196f6343a592880333dfed7f6b |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: fungus.
Question: There was more fungus under my fingernails than under my toenails because the _ were dirtier.
Answer: fingernails.
Output:
| [
"There was more fungus under my fingernails than under my toenails because the _ were cleaner."
] | task034-88cc7beda7464af298ba22b20f2d4572 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: cake.
Question: At the bakery shop, the cake was moist but the muffin was dry. The _ was fresh.
Answer: cake.
Output:
| [
"At the bakery shop, the cake was moist but the muffin was dry. The _ was old."
] | task034-8645b96b32774e6db53165513de0ce42 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: clean.
Question: I need to transfer cleaning solution from a bottle to a cup until the _ is empty.
Answer: bottle.
Output:
| [
"I need to transfer cleaning solution from a bottle to a cup until the _ is full."
] | task034-3a4c6e6d6ac04fe28e0f55fa76a59b23 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: shirt.
Question: He wore his T shirt instead of the dress shirt, because the _ was still wet.
Answer: dress shirt.
Output:
| [
"He wore his T shirt instead of the dress shirt, because the _ was already dry."
] | task034-e08e2ac364404263bff84e2f8f075953 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: writing.
Question: Writing the book was easy, but the essay turned out to be a different matter, because the material in the _ was so simple to communicate.
Answer: book.
Output:
| [
"Writing the book was easy, but the essay turned out to be a different matter, because the material in the _ was so complex to communicate."
] | task034-638625caaf1146f793d7cd227410fc33 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: paper.
Question: Dave was shopping for paper with choices of wide or college ruled so he picked the _ because the lines are spaced thinner.
Answer: college ruled.
Output:
| [
"Dave was shopping for paper with choices of wide or college ruled so he picked the _ because the lines are spaced more."
] | task034-a0dfbfe0e9c744ac9a7397db923c16ed |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: College.
Question: On the college campus, Adam threw the ball against the wall but it didn't bounce back because the _ was too deflated.
Answer: ball.
Output:
| [
"On the college campus, Adam threw the ball against the wall but it didn't bounce back because the _ was too soft."
] | task034-33108b5016fb4997a602ea908e865fea |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: acquire.
Question: As we acquired more money it would not fit in the briefcase since the _ was too large.
Answer: money.
Output:
| [
"As we acquired more money it would not fit in the briefcase since the _ was too small."
] | task034-badcbf98b5604da68dfddda1266a601a |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: Tape.
Question: On Christmas Eve, right before giving out gifts, Adam used glue instead of tape to wrap the gift because all he had was the _ .
Answer: glue.
Output:
| [
"On Christmas Eve, right before giving out gifts, Adam used glue instead of tape to wrap the gift because he could not find any the _ ."
] | task034-a5d63b95134746a3b8c43fbcd0aaa895 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: instruments.
Question: The operating nurse took the instruments off of the table and put them on the tray until the _ was empty.
Answer: table.
Output:
| [
"The operating nurse took the instruments off of the table and put them on the tray until the _ was full."
] | task034-27e311c708ea4aafab903badd5866b19 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: groomed.
Question: Jim groomed himself better for the wedding than for the party because the _ was more formal.
Answer: wedding.
Output:
| [
"Jim groomed himself better for the wedding than for the party because the _ was more casual."
] | task034-292622ab2008403dad6ba6533cfa6971 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: teenager.
Question: The teenager doesn't like going to the beach anymore, but she'll go to the pool. The _ is far away.
Answer: beach.
Output:
| [
"The teenager doesn't like going to the beach anymore, but she'll go to the pool. The _ is very close."
] | task034-91a12e443ef64a9e9141d0a95cf890f1 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: parents.
Question: Mary decided to give her parents a cruise instead of a party, because the _ was something they hadn't been on in a while.
Answer: cruise.
Output:
| [
"Mary decided to give her parents a party instead of a cruise, because the _ was something they hadn't had in a while."
] | task034-5181e586bd51482b9bdf48f894761a31 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: seeds.
Question: She couldn't decide if she should plant the seeds in the flower pot or in the garden, she was leaning towards the _ because the seeds would get more protection.
Answer: flower pot.
Output:
| [
"She couldn't decide if she should plant the seeds in the flower pot or in the garden, she was leaning towards the _ because the seeds would get more sun."
] | task034-6569b3c83cd0420dbbecfcae1d5d2c88 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: tequila.
Question: She liked drinking tequila from a glass rather than a tumbler, because the _ had a weird scent.
Answer: tumbler.
Output:
| [
"She liked drinking tequila from a glass rather than a tumbler, because the _ had a pleasant scent."
] | task034-e6adb98c651f41029453481859e70a1d |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: clean.
Question: Jill had to clean both the sink and the toilet, but the _ was pretty clean.
Answer: sink.
Output:
| [
"Jill had to clean both the sink and the toilet, but the _ was pretty dirty."
] | task034-73cbdf1f087144ed9cd547d81c898f97 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: electricity.
Question: Al got solar panels and a small generator installed at his house for electricity as the _ would be his main power supply.
Answer: panels.
Output:
| [
"Al got solar panels and a small generator installed at his house for electricity as the _ would be his backup power supply."
] | task034-7e673c6cad2642d4b41f7d1b4535fac7 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: move.
Question: When it came to the big move, James preferred to use a cart instead of a wheelbarrow for moving heavy objects, because the _ was easy to move.
Answer: cart.
Output:
| [
"When it came to the big move, James preferred to use a cart instead of a wheelbarrow for moving heavy objects, because the _ was difficult to maneuver."
] | task034-7cb8c4a4331547edb746591cc0d6ff5c |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: aquarium.
Question: The child's aquarium would not fit on the desk because the _ was too narrow.
Answer: desk.
Output:
| [
"The child's aquarium would not fit on the desk because the _ was too big."
] | task034-169367ca6f084479983033eff756e7c6 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: beef.
Question: Laura decreased her fat intake by replacing avocado with beef because the _ is fattier.
Answer: avocado.
Output:
| [
"Laura decreased her fat intake by replacing avocado with beef because the _ is leaner."
] | task034-94ce3b2b007e450f9ab475d1453c3fe1 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: birdcage.
Question: The man put the birdcage on the mantel instead of the table because the _ was farther from the cat.
Answer: mantel.
Output:
| [
"The man put the birdcage on the mantel instead of the table because the _ was closer to the cat."
] | task034-e0277e2b4ae04de095af90b513f9e87f |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: raise.
Question: The crane could not raise the ship, but could bring up the boat, as the _ was heavy for its size.
Answer: ship.
Output:
| [
"The crane could not raise the ship, but could bring up the boat, as the _ was light for its size."
] | task034-a8cd66873f11415bbdbae810fc3896fc |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: bromelain.
Question: You can take a supplement or eat pineapple to get bromelain but the _ may be cheaper.
Answer: pineapple.
Output:
| [
"You can take a supplement or eat pineapple to get bromelain but the _ may be more expensive."
] | task034-2d83f909df41423b82eda02c2d2c8c5a |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: Waste.
Question: I wanted to reduce waste from trash so I had to choose between recycling or re-purposing. The _ gets me money.
Answer: recycling.
Output:
| [
"I wanted to reduce waste from trash so I had to choose between recycling or re-purposing. The _ gets me utility."
] | task034-e915ec1df07b4ee899616a8ac40a32fb |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: swelling.
Question: Tom wanted to avoid swelling of his feet in the hot weather so he drank water but the _ was not enough.
Answer: water.
Output:
| [
"Tom wanted to avoid swelling of his feet in the hot weather so he drank water but the _ was too intense."
] | task034-b6823972054e43fcbb70ea497e4adbbb |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: phones.
Question: The teacher made the students fork up their phones so they would pay attention in class because the _ is important.
Answer: class.
Output:
| [
"The teacher made the students fork up their phones so they would pay attention in class because the _ are distracting."
] | task034-87ea8c678892440ea62f720e83f168b1 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: fishing.
Question: My grandfather gave me his old fishing reel to go with my new rod so the _ was antique.
Answer: reel.
Output:
| [
"My grandfather gave me his old fishing reel to go with my new rod so the _ was modern."
] | task034-73d92965827145059f9c90e5ad65ed46 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: excessively.
Question: The girl was working out excessively, but isn't eating healthy. That's because the _ aspect of healthy eaten was more fun for her.
Answer: working out.
Output:
| [
"The girl was working out excessively, but isn't eating healthy. That's because the _ aspect of healthy eaten was more difficult for her."
] | task034-14a6617c905c4fb4a611da7a479bd673 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: stops.
Question: The block stops the car from rolling downhill, but the slab doesn't, because the _ is too light.
Answer: slab.
Output:
| [
"The block stops the car from rolling downhill, but the slab doesn't, because the _ is heavy enough."
] | task034-e98262aa77104fe19f3d91bcd235d2f0 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: job.
Question: The job at the school paid more than the job at the factory because you needed a degree to work at the _ .
Answer: school.
Output:
| [
"The job at the school paid more than the job at the factory because you needed no degree to work at the _ ."
] | task034-c98c414923cf469ba470038f2cb67d14 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: trashcan.
Question: The garbage truck had taken the trashbag out of Mary’s trashcan this morning, so the _ is now disposed.
Answer: trashbag.
Output:
| [
"The garbage truck had taken the trashbag out of Mary’s trashcan this morning, so the _ is now empty."
] | task034-2593f528b176424eaa43187d6a97c954 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: gel.
Question: The gel James applied on his palm disappeared after rubbing them together. The _ is broad.
Answer: palm.
Output:
| [
"The gel James applied on his palm disappeared after rubbing them together. The _ is small."
] | task034-332fff555b924992964a76c2f86f738f |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: courage.
Question: Franklin needed courage to drive his car after the truck accident, because the _ still haunted him now.
Answer: truck.
Output:
| [
"Franklin needed courage to drive his car after the truck accident, but the _ didn't bother him now."
] | task034-2c47631ecc284446b63e4316f9b7269c |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: space.
Question: James was finding a space on the couch for the box to be placed but he could not find it because the _ is big.
Answer: box.
Output:
| [
"James was finding a space on the couch for the box to be placed but he could not find it because the _ is small."
] | task034-f18ab8ce18de47a29f3ca5dd4d14fc36 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: adult.
Question: The adult was unable to put the cookie sheet in the oven, because the _ was too small.
Answer: oven.
Output:
| [
"The adult was unable to put the cookie sheet in the oven, because the _ was too large."
] | task034-13bc02a41ef344f1aae17f5e41b4bbd6 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: mortgage.
Question: The wealthy family had to take out a mortgage on the condo but paid cash for the house. The _ 's price was outrageous in comparison to the other.
Answer: condo.
Output:
| [
"The wealthy family had to take out a mortgage on the condo but paid cash for the house. The _ 's price was modest in comparison to the other."
] | task034-bd5c99c4210447dbb28418dd63049adb |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: job.
Question: Alex interviewed for a job at the Diner. The Diner offered him a _ but it didn't pay enough.
Answer: job.
Output:
| [
"Alex interviewed for a job at the Diner. The Diner offered him a job but the _ didn't pay enough."
] | task034-42e0a60947ff46688a09ca49cdd2b0cb |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: box.
Question: The box was still visible after James tried his best to manage the wrapper on it. Jame should have used the _ that is big.
Answer: wrapper.
Output:
| [
"The box was still visible after James tried his best to manage the wrapper on it. Jame should have used the _ that is small."
] | task034-e4ac636d906b4ae6896fe746df36edaa |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: car.
Question: My friend Joe liked his new car better than his old truck, because the _ was more expensive.
Answer: car.
Output:
| [
"My friend Joe liked his new car better than his old truck, because the _ was more cheap."
] | task034-dc625857bb5248e48d79091679a2cfb0 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: customize.
Question: The boy decided the customize the sunglasses instead of the shoes because the _ were common.
Answer: sunglasses.
Output:
| [
"The boy decided the customize the sunglasses instead of the shoes because the _ were already unique."
] | task034-1f2c16b9926c41adae5bf93eb6e1280a |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: hair.
Question: James hair got burnt without the ribbon used to tie it because the _ is flammable.
Answer: hair.
Output:
| [
"James hair got burnt without the ribbon used to tie it because the _ is inflammable."
] | task034-1b0554fcbb98497ab908f03b8b977cd5 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: cilantro.
Question: A large amount of cilantro was thrown into the fat burrito because the _ was flavorful.
Answer: cilantro.
Output:
| [
"A large amount of cilantro was thrown into the fat burrito because the _ was flavorless."
] | task034-5bbdbace3dc74658abdb671f620432f6 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: door.
Question: James cannot get the front door open, hence he tried to remove the screw because the _ is loose.
Answer: screw.
Output:
| [
"James cannot get the front door open, hence he tried to remove the screw because the _ is tight."
] | task034-74d6d3f00b1946fbb4eb8d9d3921abca |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: meal.
Question: No one could eat the meal served at the even and the juice finished in little time because the _ is tasteless.
Answer: meal.
Output:
| [
"No one could eat the meal served at the even and the juice finished in little time because the _ is tasty."
] | task034-99beed1e987644f588159f83c3dfe2cd |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: hair.
Question: She decided to buy a new brush to keep her hair straight. The _ was soft.
Answer: hair.
Output:
| [
"She decided to buy a new brush to keep her hair straight. The _ was firm."
] | task034-9451bed625e44c1f92c96d83af77391b |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: coping.
Question: After the accident, the girl had a harder time coping with the leg injury than the injury to her hand. The _ injury was severe.
Answer: leg.
Output:
| [
"After the accident, the girl had a harder time coping with the leg injury than the injury to her hand. The _ injury was mild."
] | task034-f58c33bd86d442e6a906f8dd229647a1 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: storage.
Question: The storage left on the shelf is not enough to keep the bag. The _ is big.
Answer: bag.
Output:
| [
"The storage left on the shelf is not enough to keep the bag. The _ is small."
] | task034-01aa7c2feab4454eadfb8e13d4b32d4f |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: bedroom.
Question: I dusted the dresser in the bedroom with a rag until _ was covered in dust.
Answer: rag.
Output:
| [
"I dusted the dresser in the bedroom with a rag until _ was free of dust."
] | task034-dc2d9a7307664baf96eb0a00e08fc90b |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: avoid.
Question: I avoided the live wire but not the fence it ran along because the _ was dangerous to touch.
Answer: wire.
Output:
| [
"I avoided the live wire but not the fence it ran along because the _ was safe to touch."
] | task034-c9317d95bac94b12980abcfb490e3a07 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: stronger.
Question: The metal and glass containers were both stronger than the plastic ones, but the _ was harder to clean.
Answer: metal.
Output:
| [
"The metal and glass containers were both stronger than the plastic ones, but the _ was easier to clean."
] | task034-89ca9f37364148d2b31e8dde91207696 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: Pug.
Question: The Pug of Tristan never plays with his ball, however loves his bone, therefore the _ must be tedious to him.
Answer: ball.
Output:
| [
"The Pug of Tristan never plays with his ball, however loves his bone, therefore the _ must be gripping to him."
] | task034-68d92ccfbc8a4cd4a70faf24adf998e9 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: sleepover birthday party.
Question: Luke cleaned entire house with a vacuum cleaner for sleepover birthday party instead of a bloom because the _ will do a good and complete job.
Answer: vacuum cleaner.
Output:
| [
"Luke cleaned entire house with a vacuum cleaner for sleepover birthday party instead of a bloom because the _ will do a sloppy and incomplete job."
] | task034-54e22032d756484dbdc4d637c39f7322 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: shrinking.
Question: The pants kept shrinking in the wash, but the socks weren't affected, because the _ were made of malleable material.
Answer: pants.
Output:
| [
"The pants kept shrinking in the wash, but the socks weren't affected, because the _ were made of solid material."
] | task034-f32fca9df4c54ea58028b8bbb3d850dd |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: coffee.
Question: Tim decided to swap coffee with tea as he has read that tea contributes towards losing weight. The _ is the best option for losing weight.
Answer: tea.
Output:
| [
"Tim decided to swap coffee with tea as he has read that tea contributes towards losing weight. The _ is not so good for losing weight."
] | task034-6592ebfb4a9642deaa934d68179c514c |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: book.
Question: Since I took the books out of the box and put them into the storage chest, the _ became very light.
Answer: box.
Output:
| [
"Since I took the books out of the box and put them into the storage chest, the _ became very heavy."
] | task034-1dfb5a5655db440198cb4ca47dbbe34a |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: blood.
Question: When donating blood, I don't look at the needle until I get my cookie, and it's always great to see that the _ is quite large.
Answer: cookie.
Output:
| [
"When donating blood, I don't look at the needle until I get my cookie, and it's always great to see that the _ is quite small."
] | task034-e37a141924b445308d16fb930cc53538 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: water.
Question: James poured himself a glass of water from the bottle until he realized that there is nothing left in it. The _ is big.
Answer: glass.
Output:
| [
"James poured himself a glass of water from the bottle until he realized that there is nothing left in it. The _ is small."
] | task034-c9aceffab1c2411f8db89874e0329256 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: refrigerator.
Question: Mary replaced the gasket but not the compressor of her refrigerator, because the _ was already worn.
Answer: gasket.
Output:
| [
"Mary replaced the gasket but not the compressor of her refrigerator, because the _ is still good."
] | task034-fcf0ad9598f44fa284e6fe7bfa6391e9 |
Definition: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Positive Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because _ is too small.
Positive Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ looks great on her.
Negative Example 1 -
Input: Context word: fit.
Question: The trophy doesn't fit into the brown suitcase because _ is too large.
Answer: trophy.
Output: The trophy doesn't fit into the brown suitcase because there were a lot of other eseential items and jewlleries to pack and _ is too large.
Negative Example 2 -
Input: Context word: trade.
Question: Grace was happy to trade me her sweater for my jacket. She thinks _ looks dowdy on her.
Answer: sweater.
Output: Grace was happy to trade me her sweater for my jacket. She thinks _ is colored purple.
Now complete the following example -
Input: Context Word: bow tie.
Question: For the wedding, Kyle chose not to wear a necktie, but went with a bow tie instead. He finds the _ to be standard.
Answer: necktie.
Output:
| [
"For the wedding, Kyle chose not to wear a necktie, but went with a bow tie instead. He finds the _ to be distinctive."
] | task034-1fa343d8174448dab28a0971f9a09c27 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.