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: Spinach.
Question: Amber picked a wood container over a clay container to plant her spinach, _ allows optimal food growth.
Answer: wood container.
Output:
| [
"Amber picked a wood container over a clay container to plant her spinach, _ suppresses optimal food growth."
] | task034-ad38bcbccde84b4a899669c6d83b3419 |
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: sick.
Question: Cream pies make Michael sick, but doughnuts make him feel good because the _ are rough on his stomach.
Answer: cream pies.
Output:
| [
"Cream pies make Michael sick, but doughnuts make him feel good because the _ are satisfying to his stomach."
] | task034-6c2cef0a124a46879d6d58b5459ce017 |
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: scalpel.
Question: The doctor had a hard time trying to cut into the Thymoma with a scalpel because the _ was thick.
Answer: Thymoma.
Output:
| [
"The doctor had a hard time trying to cut into the Thymoma with a scalpel because the _ was small."
] | task034-f5f580e31ed24da491430041542b1271 |
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: bulldozer.
Question: I wanted to learn to operate a bulldozer this summer at the mine, so I asked my uncle to show me how to operate the _ .
Answer: bulldozer.
Output:
| [
"I wanted to learn to operate a bulldozer this summer at the mine, so I asked my uncle to show me at the _ ."
] | task034-dbb35fc535444279a4a3f0947b0b3322 |
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: He tried to avoid the buildings and vehicles, but the _ were harder to avoid as his speed increased.
Answer: vehicles.
Output:
| [
"He tried to avoid the buildings and vehicles, but the _ were easier to avoid as his speed increased."
] | task034-fe399d360e8447e8ab37d5c8f5b5a10e |
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: paternity.
Question: Declaring paternity is important for official records like medical records but not credit records because the _ are restricted.
Answer: medical records.
Output:
| [
"Declaring paternity is important for official records like medical records but not credit records because the _ are open."
] | task034-11e9772b717d4e4e9b6c55af02b768de |
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: I really enjoyed the meal that my mom made, but not the lunch that my dad made, because the _ was delicious.
Answer: meal.
Output:
| [
"I really enjoyed the meal that my mom made, but not the lunch that my dad made, because the _ was gross."
] | task034-23dfb136c45e494e96eaac427cc86467 |
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: seedlings.
Question: The seedlings failed to sprout in the garden so were moved to a pot, because the _ failed to provide proper nutrients.
Answer: garden.
Output:
| [
"The seedlings failed to sprout in the garden so were moved to a pot, because the _ was able to provide proper nutrients."
] | task034-7f0c851cf8ab4d22a39676398cb55b51 |
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: camping.
Question: While camping in Kansas, we had to decide whether to go north or south next, so we chose the _ because we wanted to see the aurora borealis.
Answer: north.
Output:
| [
"While camping in Kansas, we had to decide whether to go north or south next, so we chose the _ because we wanted to see the equator."
] | task034-1a8ce3ad9f0d4ddf887bd7a0c5c4a55f |
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: diamond.
Question: The diamond extracted from the mine could not be sold, but the one crafted in the laboratory sold well, because the _ produced rough diamonds.
Answer: mine.
Output:
| [
"The diamond extracted from the mine could not be sold, but the one crafted in the laboratory sold well, because the _ produced polished diamonds."
] | task034-8b7eae62174f45169d5eb38b9c3524c7 |
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: portfolio.
Question: The manager at the ad agency was impressed by Mary's portfolio. The _ needed creativity.
Answer: ad agency.
Output:
| [
"The manager at the ad agency was impressed by Mary's portfolio. The _ was creative."
] | task034-34b5869f334143c1a7499a79f4c76472 |
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: bullies.
Question: The _ are very good at dealing with bullies but the gangsters are not because the courts are very moral.
Answer: courts.
Output:
| [
"The courts are very good at dealing with bullies but the _ are not because the gangsters are very amoral."
] | task034-e0f4e0f762b44968a36f5d4fa4765ef9 |
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: purchase.
Question: At the register I decided to purchase the name brand over the store brand as the _ was cheaper.
Answer: name brand.
Output:
| [
"At the register I decided to purchase the name brand over the store brand as the _ was more expensive."
] | task034-26fd3d7229114183a67ff760193aa23c |
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: gold.
Question: Some ounces of gold was sold in exchange for a yacht. I never know the _ is that cheap.
Answer: gold.
Output:
| [
"Some ounces of gold was sold in exchange for a yacht. I never know the _ is that expensive."
] | task034-b9a47f50805b4639bd505bf6feb4d0e5 |
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: article.
Question: The article could not fit on the sheet of paper that was given to him because the _ was too long.
Answer: article.
Output:
| [
"The article could not fit on the sheet of paper that was given to him because the _ was too short."
] | task034-d93a7c5b727e43ddb54dd86c9e74e663 |
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: colder.
Question: It was colder in the trailer than it was in the house because the heater in the _ was broken.
Answer: trailer.
Output:
| [
"It was colder in the trailer than it was in the house because the heater in the _ was turned on."
] | task034-d78153dc0cea4fe88c17c74f368e8483 |
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: teapot.
Question: I thought that I had enough water for the teapot but I didn't because the _ was so little.
Answer: water.
Output:
| [
"I thought that I had enough water for the teapot but I didn't because the _ was so huge."
] | task034-66c2f73570d84431b35a25cd9ca50716 |
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: write.
Question: James chose to write his notes in a notebook instead of a tablet despite finding the weight of the _ very convenient.
Answer: tablet.
Output:
| [
"James chose to write his notes in a notebook instead of a tablet despite finding the weight of the _ very inconvenient."
] | task034-d65614b82a0f4d45b7b0535dc0e3fa7f |
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: crystal.
Question: They decided to buy crystal goblets instead of pewter tankards because the _ were old-fashioned.
Answer: tankards.
Output:
| [
"They decided to buy crystal goblets instead of pewter tankards because the _ were modern."
] | task034-52dbbe49ff984709954b1dd0f0006bda |
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: popcorn.
Question: They grabbed some popcorn when they got to the movie theater because they wanted to snack on the _ .
Answer: popcorn.
Output:
| [
"They grabbed some popcorn when they got to the movie theater because they wanted to snack during the _ ."
] | task034-58b6ec4134cf4ea490ff5019747bb732 |
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: message.
Question: The message sent by email arrived after the one sent by text because the _ was delayed.
Answer: email.
Output:
| [
"The message sent by email arrived after the one sent by text because the _ was immediate."
] | task034-1f2191a222fc4671b496cd2fac1a89b2 |
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: He wanted to replace his skin that he was wearing with an armor, because the _ was damagable.
Answer: skin.
Output:
| [
"He wanted to replace his skin that he was wearing with an armor, because the _ was impenetrable."
] | task034-656a754f33dc4d709dd58d1ba2b751d1 |
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: tough.
Question: My mother forgot about the Sunday roast in the oven and it became too tough to eat, because the _ dried out.
Answer: roast.
Output:
| [
"My mother forgot about Sunday roast in the oven and it became too tough to eat, because the _ dried it out."
] | task034-e2ac0ac8bb334702907b10a4d9e4ac85 |
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: turtle.
Question: Hamza's pet turtle liked to be in the water dish instead of the pond, because the _ was safer.
Answer: dish.
Output:
| [
"Hamza's pet turtle liked to be in the water dish instead of the pond, but the _ was larger."
] | task034-b52066a8a55d4657a8c5d095b6c850a3 |
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: church.
Question: The church decided the replace the pews but keep the candles, because the _ were old.
Answer: pews.
Output:
| [
"The church decided the replace the pews but keep the candles, because the _ were new."
] | task034-5895684baf7a4637b8e4194594b50a7c |
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: contract.
Question: The contract did not allow the group to use the auditorium, but they could use the stadium. The _ was a restricted space.
Answer: auditorium.
Output:
| [
"The contract did not allow the group to use the auditorium, but they could use the stadium. The _ was an available space."
] | task034-1d33768f94ba4d2da3b4ba59f1aa1c05 |
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: The teacher's old book could not be put on the shelf because the _ was too flimsy.
Answer: shelf.
Output:
| [
"The teacher's old book could not be put on the shelf because the _ was too heavy."
] | task034-5212cd91ee4849f9802b173abcd0fe45 |
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: shoes.
Question: She had to leave the hiking boots at home and take the dress shoes because this trip was business so the _ were unnecessary.
Answer: boots.
Output:
| [
"She had to leave the hiking boots at home and take the dress shoes because this trip was business so the _ were necessary."
] | task034-9ef6a3e378fc41529febf31e31467fbe |
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: online.
Question: Asha checked the header and the copyright information to know if an online store was legitimate, but the _ did not look professional to her.
Answer: header.
Output:
| [
"Asha checked the header and the copyright information to know if an online store was legitimate, but the _ did not look legal to her."
] | task034-f8832db13e784962a4ec820b57a4d1db |
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: friends.
Question: I had to choose between seeing my friends at the restaurant or at the mall, we went to the _ because there's more to do.
Answer: mall.
Output:
| [
"I had to choose between seeing my friends at the restaurant or at the mall, we went to the _ because there's more to choose to eat."
] | task034-da9a16e03f8c4175860820fe23b07df3 |
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: babies.
Question: The babies cried in the playroom but were happy in the nursery, as the _ was a scary place for them.
Answer: playroom.
Output:
| [
"The babies cried in the playroom but were happy in the nursery, as the _ was a pleasant place for them."
] | task034-519136b75142456f96542eaa072f7faa |
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: bought.
Question: Jack passed on the apples and bought the grapes instead because the _ cost much more.
Answer: apples.
Output:
| [
"Jack passed on the apples and bought the grapes instead because the _ cost much less."
] | task034-f07a6799237b4c969d04b7fe2874cd0d |
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: girlfriend.
Question: Joey used communication in the form of text to break off their relationship with his girlfriend but the _ was too complicated.
Answer: relationship.
Output:
| [
"Joey used communication in the form of text to break off their relationship with his girlfriend but the _ was too easy."
] | task034-2dd6e0a351b14b439db8b21d115ba151 |
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: bunny.
Question: The bunny spat out the carrots but munched happily on the beans, since the _ were unappetizing.
Answer: carrots.
Output:
| [
"The bunny spat out the carrots but munched happily on the beans, since the _ were delicious."
] | task034-00a36d6c44444248899e94b5ebfcbd85 |
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: pet.
Question: Becky was hired as a pet sitter for the weekend, and she slept on the floor instead of the bed because the _ smelled funny.
Answer: bed.
Output:
| [
"Becky was hired as a pet sitter for the weekend, and she slept on the floor instead of the bed because the _ smelled better."
] | task034-3ca1981d96c34238beeb5891fa1474b5 |
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: drums.
Question: The set of electronic drums didn't fit on the stage because the _ is too big.
Answer: drums.
Output:
| [
"The set of electronic drums didn't fit on the stage because because the _ is too small."
] | task034-e435ec1ba7324a4f8015a7a089cf5518 |
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: hook.
Question: I threw the hook to lasso the shirt, but I couldn't get it because the _ was too dull.
Answer: hook.
Output:
| [
"I threw the hook to lasso the shirt, but I couldn't get it because the _ was too far."
] | task034-be18fcb7f96f4eaaa3c8bdbf7e5e2fad |
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: plant.
Question: I need to plant plants but I need to move the dirt from the hole to the pot. I move the dirt until the _ is empty.
Answer: hole.
Output:
| [
"I need to plant plants but I need to move the dirt from the hole to the pot. I move the dirt until the _ is full."
] | task034-62fab584b83e44f0b9c6243a97d0699d |
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: horse.
Question: The horse had stiff legs so its owner called the vet, who worried it had stifles. The vet was wrong and the _ were fine.
Answer: legs.
Output:
| [
"The horse had stiff legs so its owner called the vet, who worried it had stifles. The vet was right and it was the _ ."
] | task034-fd828949a8624da29af1e9ca66e0c1a2 |
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: boomerang.
Question: The boomerang was preferred to the frisbee by the children because the _ was more unique.
Answer: boomerang.
Output:
| [
"The boomerang was preferred to the frisbee by the children because the _ was more mundane."
] | task034-e0676f81942145299f50e7aeb5983fa9 |
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: ears.
Question: Fatima got her ears pierced with gold earrings rather than silver, because the _ looked expensive.
Answer: gold.
Output:
| [
"Fatima got her ears pierced with gold earrings rather than silver, because the _ looked cheap."
] | task034-87b8b26a01f0435db7824f41cf3456a1 |
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: lawn.
Question: Josh was working on his lawn and had to deal with the weeds and the flowers. He worked on the _ first because they were ugly.
Answer: weeds.
Output:
| [
"Josh was working on his lawn and had to deal with the weeds and the flowers. He worked on the _ first because they were pretty."
] | task034-f278e897756e4e83821012cc56d79766 |
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: horse.
Question: I had to shovel hay until all the horse stalls were full. Luckily the _ was pretty light.
Answer: hay.
Output:
| [
"I had to shovel hay until all the horse stalls were full. Luckily the _ was pretty sturdy."
] | task034-5f97227378e04e39ab453ed40cfce9b2 |
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: legal.
Question: Both balloons and knives were legal locally, but many found the _ to be safe.
Answer: balloons.
Output:
| [
"Both balloons and knives were legal locally, but many found the _ to be dangerous."
] | task034-6edaf9bf1dc542838947035afac12de1 |
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: eye.
Question: Richard has an eye strain because he works in a dark room close to the computer whole day. The _ is very bright.
Answer: computer.
Output:
| [
"Richard has an eye strain because he works in a dark room close to the computer whole day. The _ is very dark."
] | task034-afef2038c66d4d94a71f0221ae13d7ef |
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: algae.
Question: There was more algae in the pond than there was in the pool because the _ was cleaner.
Answer: pool.
Output:
| [
"There was more algae in the pond than there was in the pool because the _ was dirtier."
] | task034-1ff389b92e5a4911a7aa7fb455f6b939 |
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: Josiah had a choice of two coffee flavors, mocha or caramel. The _ was too bitter for him.
Answer: mocha.
Output:
| [
"Josiah had a choice of two coffee flavors, mocha or caramel. The _ was too sweet for him."
] | task034-eca97d4a770c48aba936b8e99ca15258 |
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: travel.
Question: Chase was a mile away and Bank of America was 3 miles away. Tom traveled to the _ second because it was farther.
Answer: Bank of America.
Output:
| [
"Chase was a mile away and Bank of America was 3 miles away. Tom traveled to the _ first because it was closer."
] | task034-e4d2ef5efc6f4d1e98d700e888d56240 |
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: field.
Question: The field was lacking in flowers but had plenty of weeds, as it was lacking in the nutrients needed by the _ .
Answer: flowers.
Output:
| [
"The field was lacking in flowers but had plenty of weeds, as it was abundant in the nutrients needed by the _ ."
] | task034-14ce1b8362fe4b92879723dcf7a073aa |
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: lettuce.
Question: The lettuce tasted better than the cabbage so I used the _ in the side dish when I cooked.
Answer: lettuce.
Output:
| [
"The lettuce tasted worse than the cabbage so I used the _ in the side dish when I cooked."
] | task034-989ebf8eeda043cfb5f870d049d00fa6 |
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: story.
Question: The boy liked the story more than the movie because the _ was more interesting.
Answer: story.
Output:
| [
"The boy liked the story more than the movie because the _ was more boring."
] | task034-81f92cd8a25741d0a7842af4a930e04b |
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: add.
Question: I needed to add two more teaspoons of sweetener to the coffee because the _ was far too strong.
Answer: coffee.
Output:
| [
"I needed to add two more teaspoons of sweetener to the coffee because the _ was far too weak."
] | task034-6b5fae5a36844e2685b59ad95c713a90 |
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: He couldn't decide between the guitar and the harmonica but he ended up choosing the harmonica because the _ was cheaper.
Answer: harmonica.
Output:
| [
"He couldn't decide between the guitar and the harmonica but he ended up choosing the harmonica because the _ was more expensive."
] | task034-5e9c6d43d0304fa2968aca5c1d5ed75e |
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: motor.
Question: I get my motor oil changed a few times a year but only get brakes done once. The _ take longer to wear out.
Answer: brakes.
Output:
| [
"I get my motor oil changed a few times a year but only get brakes done once. The _ doesn't take long to wear out."
] | task034-a75ba7a07da4468085ca54c498d671b0 |
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: chopstick.
Question: I was able to use the fork easier than the chopstick because I grew up in America and the _ is more foreign.
Answer: chopstick.
Output:
| [
"I was able to use the fork easier than the chopstick because I grew up in America and the _ is more natural."
] | task034-5267e3baa5f34669949b7dcb5ad91edb |
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: hospital.
Question: I decided to go to the hospital and not the clinic because the _ was more comfortable and relaxing.
Answer: hospital.
Output:
| [
"I decided to go to the hospital and not the clinic because the _ was uncomfortable and stressful."
] | task034-ea2cc54a592e4e50845da360131293da |
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: scale.
Question: The woman's kitchen scale failed to measure the white powder because the _ was inaccurate.
Answer: scale.
Output:
| [
"The woman's kitchen scale failed to measure the white powder because the _ was precise."
] | task034-678dcaab3b5d42869d63611947743460 |
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: weight.
Question: I used the weight bench more than I used the treadmill because the _ was never available.
Answer: treadmill.
Output:
| [
"I used the weight bench more than I used the treadmill because the _ was always available."
] | task034-84e3c8d326f14330beb6ead89a42dcd7 |
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: flask.
Question: Rick poured the vodka from a bottle into his flask until the _ was empty.
Answer: bottle.
Output:
| [
"Rick poured the vodka from a bottle into his flask until the _ was full."
] | task034-6b246059672d4b65bc4f8f27dc4ec5a1 |
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: ipad.
Question: The ipad had the book pre loaded but not the film as the _ was completely free.
Answer: book.
Output:
| [
"The ipad had the book pre loaded but not the film as the _ was very expensive."
] | task034-7cee3459f8064b2f98e1f652f182bde7 |
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: safe.
Question: When given the choice between the road and alley, I play it safe by taking the _ because it is well lit.
Answer: road.
Output:
| [
"When given the choice between the road and alley, I play it safe by avoiding the _ because it is dark."
] | task034-30df9741d0e24208ab24393f66d5455e |
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: measure.
Question: James needed another tape measure to completely measure the circumference of the tube. The _ is short.
Answer: measure.
Output:
| [
"James needed another tape measure to completely measure the circumference of the tube. The _ is wide."
] | task034-965921d619d140fe92e0ef636402eafb |
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: necklace.
Question: The antique ring, bracelet, and necklace that Jill gave Patty was stunning to look at. By the same token the _ still glistened.
Answer: bracelet.
Output:
| [
"The antique ring, bracelet, and necklace that Jill gave Patty was stunning to look at. By the same token the _ still darkened."
] | task034-a3d22b408f81441e9c83b11b8b93bf97 |
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: We tried to put all of the seeds into two jars, but we couldn't because the _ were too many.
Answer: seeds.
Output:
| [
"We tried to put all of the seeds into two jars, but we couldn't because the _ were too small."
] | task034-bdcdb0a42044477798be0050b8dfab7c |
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: spanish.
Question: Jessica went to class to learn Spanish to travel to another country. She went to the _ for teaching.
Answer: class.
Output:
| [
"Jessica went to class to learn Spanish to travel to another country. She went to the _ for vacation."
] | task034-7834ccdaae21418fa60ba3ccbcda375d |
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: breath.
Question: They had the mint and cinnamon breath fresheners today, but the _ wasn't the one she liked.
Answer: mint.
Output:
| [
"They had the mint and cinnamon breath fresheners today, but the _ was the only one she liked."
] | task034-79fdac719a50430a9d8c0e2e7be4265e |
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: cutting board.
Question: Jim was shopping for a new cutting board made of glass or wood. He chose the _ because it was less porous.
Answer: glass.
Output:
| [
"Jim was shopping for a new cutting board made of glass or wood. He chose the _ because it was less breakable."
] | task034-8d0d4a17ebed4970a8c9dfb4a5d76725 |
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: donation.
Question: His donation towards clearing the debt of the charity didn't help because the _ was too small.
Answer: donation.
Output:
| [
"His donation towards clearing the debt of the charity didn't help because the _ was too large."
] | task034-653731eb65c4442d95687733d2273fb8 |
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: liquor.
Question: I tried to pour the liquor into the bottle, but it wouldn't fit because the _ was too much.
Answer: liquor.
Output:
| [
"I tried to pour the liquor into the bottle, but it wouldn't fit because the _ was too short."
] | task034-75f1881b45484eeb87743b4f2c720c2a |
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: comfortable.
Question: Jimmy was more comfortable repairing the computer than the car because the _ was smaller.
Answer: computer.
Output:
| [
"Jimmy was more comfortable repairing the computer than the car because the _ was larger."
] | task034-f1ae77bcdc764bf9bedb88efa81421b8 |
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: cramp.
Question: The runner wasn't able to finish the half marathon after being hit with a crippling cramp in his thigh, because the _ was painful.
Answer: cramp.
Output:
| [
"The runner wasn't able to finish the half marathon after being hit with a crippling cramp in his thigh, because the _ was timed."
] | task034-07408fa772234093b556e4804ce69dbb |
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: act.
Question: At the theater last night, the act went on until 10 pm unlike the skit, so the _ was more in depth.
Answer: act.
Output:
| [
"At the theater last night, the act went on until 10 pm unlike the skit, so the _ was more shallow."
] | task034-b38732a4d79a4afd8f1b3b7f8022dc10 |
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: tablet.
Question: John broke the tablet when he placed the box on it not knowing the _ is fragile.
Answer: tablet.
Output:
| [
"John broke the tablet when he placed the box on it not knowing the _ is heavy."
] | task034-02b0dc9672fe485ca4779c074a992c2b |
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: debt.
Question: She tried to pay off her debt with a regular monthly payment, but the _ was just too high.
Answer: debt.
Output:
| [
"She tried to pay off her debt with a regular monthly payment, but the _ was just too low."
] | task034-c14ca0cbcbf642f99185f646199dc55a |
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: soccer.
Question: Before I play soccer, I prefer to eat a granola bar instead of an apple because the _ is quite filling.
Answer: granola bar.
Output:
| [
"Before I play soccer, I prefer to eat a granola bar instead of an apple because the _ is not filling."
] | task034-193d5747ca534d078b4ae4d1086a97c0 |
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: winter.
Question: In winter, he always wore a thick jacket with his combat boots. The _ was heavy.
Answer: jacket.
Output:
| [
"In winter, he always wore a thick jacket with his combat boots. The _ was rugged."
] | task034-4fbedade7d2640bc984bfa84340202aa |
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: tree.
Question: More people were sitting near the tree than near the bush because the _ offered shade.
Answer: tree.
Output:
| [
"More people were sitting near the tree than near the bush because the _ had bees."
] | task034-1250971cac2c4887a6aac23b11edd4e9 |
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: assets.
Question: The assets were more protected by the bank than the insurance company because the _ had more money.
Answer: bank.
Output:
| [
"The assets were more protected by the bank than the insurance company because the _ had less money."
] | task034-dd27875aaf2d4d8bbfe744d469eaca93 |
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: playground.
Question: With nothing for kids to do in the neighborhood, when voting for a new playground or a tavern, Odette voted for the _ being built.
Answer: playground.
Output:
| [
"With nothing for kids to do in the neighborhood, when voting for a new playground or a tavern, Odette voted against the _ being built."
] | task034-2e041497ee074e22b52ba96838109fb7 |
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: shrink.
Question: No matter how much we tried to shrink it, the stick could not fit through the hole, as the _ was too wide.
Answer: stick.
Output:
| [
"No matter how much we tried to shrink it, the stick could not fit through the hole, as the _ was too narrow."
] | task034-21b8017dd872477fb097aa4bea52210e |
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: badminton.
Question: Badminton was a lot easier for Josh than tennis because the _ court was smaller.
Answer: badminton.
Output:
| [
"Badminton was a lot easier for Josh than tennis because the _ court was larger."
] | task034-22a023e0bffd40a8bb75a8369a828e21 |
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: language.
Question: Anthony wanted to learn a new language at a university or a college. He went to the _ because it was less esteemed.
Answer: college.
Output:
| [
"Anthony wanted to learn a new language at a university or a college. He went to the _ because it was not popular."
] | task034-b1f1f9b7458c4e78a15c03f321421384 |
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: ceiling.
Question: the broom could not reach the ceiling while i was clearing the cobweb because the _ is short.
Answer: broom.
Output:
| [
"the broom could not reach the ceiling while i was clearing the cobweb because the _ is tall."
] | task034-7c33301166d741a1acfa82b84752473e |
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: hat.
Question: James removed the hat that is making his head sweat and wore a band. The _ is big.
Answer: hat.
Output:
| [
"James removed the hat that is making his head sweat and wore a band. The _ is small."
] | task034-8970f58e77e544a5a3061fe56195a094 |
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: placement.
Question: Due to their placement, the roadster had a disadvantage in the race against the coupe. The _ started the road race behind the other car.
Answer: roadster.
Output:
| [
"Due to their placement, the roadster had a disadvantage in the race against the coupe. The _ started the road race leading the other car."
] | task034-089852409206471daa557976b04ebc2e |
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: steak.
Question: The steak needed a hearty pinch of salt after it was cooked because the _ tasted bland.
Answer: steak.
Output:
| [
"The steak needed a hearty pinch of salt after it was cooked because the _ tasted savory."
] | task034-f4712948f5fa4e899565e51889d971d8 |
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: sugar.
Question: His nephew usually preferred to eat the sugar more than the grapefruit seeing that the _ was bitter.
Answer: grapefruit.
Output:
| [
"His nephew usually preferred to eat the sugar more than the grapefruit seeing that the _ was sweet."
] | task034-77b2b6332d5b42e195ca1575bff857c9 |
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: ride.
Question: Tom has two rides an old car that he loves and he just bough a new cool bike as well. The _ is old.
Answer: car.
Output:
| [
"Tom has two rides an old car that he loves and he just bough a new cool bike as well. The _ is new and cool."
] | task034-76e69f4374a84f6a8039cacbaa2c0239 |
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: developing.
Question: As the plant was developing, it got too big for the cup and was moved to the jug, making the _ full.
Answer: jug.
Output:
| [
"As the plant was developing, it got too big for the cup and was moved to the jug, making the _ empty."
] | task034-3a7c5112939248d3a641d9c413c66832 |
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: crown.
Question: Matt tried to pick up the crown and put it on the table behind the car, but the _ was too small.
Answer: table.
Output:
| [
"Matt tried to pick up the crown and put it on the table behind the car, but the _ was in the way."
] | task034-8069ff1d4ccf4fe98456989c9d36592c |
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: girl.
Question: The girl chose to eat the apple instead of the fries because the _ were greasier.
Answer: fries.
Output:
| [
"The girl chose to eat the apple instead of the fries because the _ was healthier."
] | task034-691dc4c5822c4997b302e9c26602aba2 |
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 slid down James hand when he was wearing a glove, hence he removed it and lift it again. The _ is slippery.
Answer: glove.
Output:
| [
"The box slid down James hand when he was wearing a glove, hence he removed it and lift it again. The _ is dry."
] | task034-136bf83789804ce499f984aa2b2fe2e2 |
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: pool.
Question: He had to move the pool table from the bedroom to the living room, because the _ was smaller.
Answer: bedroom.
Output:
| [
"He had to move the pool table from the bedroom to the living room, because the _ was larger."
] | task034-d7225dc8b1b0489d8a026ddb92f63967 |
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: toddler.
Question: The toddler is just starting to eat and doesn't like eating blueberries when compared to grapes. The _ are overpowering to him.
Answer: blueberries.
Output:
| [
"The toddler is just starting to eat and doesn't like eating blueberries when compared to grapes. The _ are ideal to him."
] | task034-ad6efb847ecb4dfd881b0978fc45a7df |
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: roses.
Question: The roses wilted in the garden so we moved them to the greenhouse, because the _ had poor conditions for the flower's growth.
Answer: garden.
Output:
| [
"The roses wilted in the garden so we moved them to the greenhouse, because the _ had ideal conditions for the flower's growth."
] | task034-6789c0e415544372a5aef7c0035b5554 |
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: routine.
Question: Jeff established a routine after he quit his job and became a NEET because the _ was insecure.
Answer: NEET.
Output:
| [
"Jeff established a routine after he quit his job and became a NEET because the _ was secure."
] | task034-d82ca032d39b48e2b669bfaa957bfb9f |
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: casket.
Question: There was an awkward situation during the service, the casket did not fit in a crypt, the _ was too short.
Answer: crypt.
Output:
| [
"There was an awkward situation during the service, the casket did not fit in a crypt, the _ was too long."
] | task034-b7cbd63e1e264f3eadd91c50e8f290b8 |
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: share.
Question: James cannot share the space on the couch with the box placed on the table because the _ is big.
Answer: box.
Output:
| [
"James cannot share the space on the couch with the box placed on the table because the _ is small."
] | task034-1a1c019efcfd477a8cea14c6c022a04b |
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: Squirrel.
Question: After Jessica found the injured squirrel she decided to put it in a cage instead of a shoe box, because the _ was larger.
Answer: cage.
Output:
| [
"After Jessica found the injured squirrel she decided to put it in a cage instead of a shoe box, because the _ was smaller."
] | task034-c5ae80248c404781b442038fba30b23a |
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: preparing.
Question: she was preparing for the wedding and the test, but she felt less prepared for the _ tomorrow.
Answer: test.
Output:
| [
"she was preparing for the wedding and the test, but she felt more prepared for the _ tomorrow."
] | task034-2fbc115d0fc9444dbac8b5123e0530c8 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.