Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
csv
Languages:
English
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
license: cc-by-nc-sa-4.0
|
|
|
|
| 2 |
task_categories:
|
| 3 |
- text-classification
|
| 4 |
language:
|
|
@@ -8,19 +9,15 @@ tags:
|
|
| 8 |
- bias-detection
|
| 9 |
- robustness
|
| 10 |
size_categories:
|
| 11 |
-
- 10K<n<100K
|
| 12 |
-
|
| 13 |
-
---
|
| 14 |
-
|
| 15 |
pretty_name: RobustBiasBench
|
| 16 |
description: >
|
| 17 |
RobustBiasBench consists of 18,404 U.S. policy text excerpts manually annotated
|
| 18 |
for bias type and normative framing. The dataset is designed to evaluate the
|
| 19 |
robustness of language models in fairness-critical tasks, with both clean and
|
| 20 |
perturbed versions of the data available.
|
| 21 |
-
|
| 22 |
---
|
| 23 |
-
|
| 24 |
dataset_format:
|
| 25 |
file_type: csv
|
| 26 |
fields:
|
|
|
|
| 1 |
license: cc-by-nc-sa-4.0
|
| 2 |
+
---
|
| 3 |
task_categories:
|
| 4 |
- text-classification
|
| 5 |
language:
|
|
|
|
| 9 |
- bias-detection
|
| 10 |
- robustness
|
| 11 |
size_categories:
|
| 12 |
+
- 10K<n<100K
|
| 13 |
+
---
|
|
|
|
|
|
|
| 14 |
pretty_name: RobustBiasBench
|
| 15 |
description: >
|
| 16 |
RobustBiasBench consists of 18,404 U.S. policy text excerpts manually annotated
|
| 17 |
for bias type and normative framing. The dataset is designed to evaluate the
|
| 18 |
robustness of language models in fairness-critical tasks, with both clean and
|
| 19 |
perturbed versions of the data available.
|
|
|
|
| 20 |
---
|
|
|
|
| 21 |
dataset_format:
|
| 22 |
file_type: csv
|
| 23 |
fields:
|