Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Tasks
We study the relationship between task-agnostic intrinsic and task-specific extrinsic social bias evaluation measures for Masked Language Models (MLMs), and find that there exists only a weak correlation between these two types of evaluation measures. Moreover, we find that MLMs debiased using different methods still re-learn social biases during fine-tuning on downstream tasks. We identify the social biases in both training instances as well as their assigned labels as reasons for the discrepancy between intrinsic and extrinsic bias evaluation measurements. Overall, our findings highlight the limitations of existing MLM bias evaluation measures and raise concerns on the deployment of MLMs in downstream applications using those measures.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Debiasing Isn’t Enough! – on the Effectiveness of Debiasing MLMs and Their Social Biases in Downstream Tasks
We study the relationship between task-agnostic intrinsic and task-specific extrinsic social bias evaluation measures for MLMs, and find that there exists only a weak correlation between these two types of evaluation mea…
Debiasing Multilingual LLMs in Cross-lingual Latent Space
Debiasing techniques such as SentDebias aim to reduce bias in large language models (LLMs). Previous studies have evaluated their cross-lingual transferability by directly applying these methods to LLM representations, r…
What Changed? Investigating Debiasing Methods using Causal Mediation Analysis
Previous work has examined how debiasing language models affect downstream tasks, specifically, how debiasing techniques influence task performance and whether debiased models also make impartial predictions in downstrea…
Addressing bias in Recommender Systems: A Case Study on Data Debiasing Techniques in Mobile Games
The mobile gaming industry, particularly the free-to-play sector, has been around for more than a decade, yet it still experiences rapid growth. The concept of games-as-service requires game developers to pay much more a…
Recommendation SystemsChatGPT Based Data Augmentation for Improved Parameter-Efficient Debiasing of LLMs
Large Language models (LLMs), while powerful, exhibit harmful social biases. Debiasing is often challenging due to computational costs, data constraints, and potential degradation of multi-task language capabilities. Thi…
Data AugmentationFairness