AI·The research finds that preprocessing-based stereotype mitigation in NLP can backfire by increasing stereotyping or counter-stereotyping for some groups relative to neutral baselines.
Search
What are you looking for?
1 result for “debiasing”
Search
1 result for “debiasing”
AI·The research finds that preprocessing-based stereotype mitigation in NLP can backfire by increasing stereotyping or counter-stereotyping for some groups relative to neutral baselines.