Update README.md
Browse files
README.md
CHANGED
@@ -30,21 +30,13 @@ More information needed
|
|
30 |
## Direct Use
|
31 |
This model can be used for the task of feature extraction.
|
32 |
|
33 |
-
## Downstream Use [Optional]
|
34 |
-
|
35 |
-
More information needed.
|
36 |
-
|
37 |
## Out-of-Scope Use
|
38 |
|
39 |
The model should not be used to intentionally create hostile or alienating environments for people.
|
40 |
|
41 |
# Bias, Risks, and Limitations
|
42 |
|
43 |
-
|
44 |
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
## Recommendations
|
49 |
|
50 |
|
@@ -57,63 +49,6 @@ Users (both direct and downstream) should be made aware of the risks, biases and
|
|
57 |
The model craters note in the [Github Repository](https://github.com/SJTU-LIT/SynCSE/blob/main/README.md)
|
58 |
> We use 26.2k generated synthetic train SynCSE-partial-RoBERTa-base.
|
59 |
|
60 |
-
|
61 |
-
## Training Procedure
|
62 |
-
|
63 |
-
|
64 |
-
### Preprocessing
|
65 |
-
|
66 |
-
More information needed
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
### Speeds, Sizes, Times
|
71 |
-
|
72 |
-
More information needed
|
73 |
-
|
74 |
-
|
75 |
-
# Evaluation
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
### Factors
|
81 |
-
More information needed
|
82 |
-
|
83 |
-
### Metrics
|
84 |
-
|
85 |
-
More information needed
|
86 |
-
|
87 |
-
|
88 |
-
## Results
|
89 |
-
|
90 |
-
More information needed
|
91 |
-
|
92 |
-
|
93 |
-
# Model Examination
|
94 |
-
|
95 |
-
The model craters note in the [associated paper](https://arxiv.org/pdf/2104.08821.pdf):
|
96 |
-
|
97 |
-
|
98 |
-
# Technical Specifications [optional]
|
99 |
-
|
100 |
-
## Model Architecture and Objective
|
101 |
-
|
102 |
-
More information needed
|
103 |
-
|
104 |
-
## Compute Infrastructure
|
105 |
-
|
106 |
-
More information needed
|
107 |
-
|
108 |
-
### Hardware
|
109 |
-
|
110 |
-
|
111 |
-
More information needed
|
112 |
-
|
113 |
-
### Software
|
114 |
-
|
115 |
-
More information needed.
|
116 |
-
|
117 |
# Citation
|
118 |
|
119 |
|
@@ -135,7 +70,6 @@ More information needed
|
|
135 |
# More Information [optional]
|
136 |
More information needed
|
137 |
|
138 |
-
|
139 |
|
140 |
|
141 |
# Model Card Contact
|
|
|
30 |
## Direct Use
|
31 |
This model can be used for the task of feature extraction.
|
32 |
|
|
|
|
|
|
|
|
|
33 |
## Out-of-Scope Use
|
34 |
|
35 |
The model should not be used to intentionally create hostile or alienating environments for people.
|
36 |
|
37 |
# Bias, Risks, and Limitations
|
38 |
|
|
|
39 |
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
|
|
|
|
|
|
|
40 |
## Recommendations
|
41 |
|
42 |
|
|
|
49 |
The model craters note in the [Github Repository](https://github.com/SJTU-LIT/SynCSE/blob/main/README.md)
|
50 |
> We use 26.2k generated synthetic train SynCSE-partial-RoBERTa-base.
|
51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
# Citation
|
53 |
|
54 |
|
|
|
70 |
# More Information [optional]
|
71 |
More information needed
|
72 |
|
|
|
73 |
|
74 |
|
75 |
# Model Card Contact
|