We use cookies to improve your experience on our site.
AVID-2022-R0004
Description
Profession bias reinforcing gender stereotypes found in xlm-roberta-base, as measured on the Winobias dataset
Details
Filling in pronouns in sentences tagged with professions using xlm-roberta-base were found to be significantly biased on the Winobias dataset.
References
- Gender Bias Evaluation for Masked Language modelling: Winobias
- xlm-roberta-base on Hugging Face
- WinoBias
AVID Taxonomy Categorization
- Risk domains: Ethics
- SEP subcategories: E0101: Group Fairness
- Lifecycle stages: L05: Evaluation
Affected or Relevant Artifacts
- Developer:
- Deployer: HuggingFace
- Artifact Details:
Type Name Model xlm-roberta-base Dataset sasha/wino_bias_cloze1 Dataset sasha/wino_bias_cloze2
Other information
- Report Type: Detection
- Credits: Harry Saini, AVID; Sasha Luccioni, Hugging Face
- Date Reported: 2022-11-09
- Version: 0.1
- AVID Entry