paper-with-me

Papers

Does Debiasing Inevitably Degrade the Model Performance

2022-11-14 · Yiran Liu, Xiao Liu, Haotian Chen, Yang Yu

Gender bias in language models has attracted sufficient attention because it threatens social justice. However, most of the current debiasing methods degraded the model's performance on other tasks while the degradation mechanism is still mysterious. We propose a theoretical framework explaining the three candidate mechanisms of the language model's gender bias. We use our theoretical framework to explain why the current debiasing methods cause performance degradation. We also discover a pathway through which debiasing will not degrade the model performance. We further develop a causality-detection fine-tuning approach to correct gender bias. The numerical experiment demonstrates that our method is able to lead to double dividends: partially mitigating gender bias while avoiding performance degradation.

📄 PDF Abstract BibTeX arXiv:2211.07350

Code (0)

등록된 구현이 없습니다.

Tasks

model

Similar Papers 제목 키워드 기반

Mitigating Perception Bias: A Training-Free Approach to Enhance LMM for Image Quality Assessment

2024-11-19 · Siyi Pan, Baoliang Chen, Danni Huang, Hanwei Zhu 외

Despite the impressive performance of large multimodal models (LMMs) in high-level visual tasks, their capacity for image quality assessment (IQA) remains limited. One main reason is that LMMs are primarily trained for h…

Image CaptioningImage Quality Assessment

Large Language Model Bias Mitigation from the Perspective of Knowledge Editing

2024-05-15 · Ruizhe Chen, Yichen Li, Zikai Xiao, Zuozhu Liu

Existing debiasing methods inevitably make unreasonable or undesired predictions as they are designated and evaluated to achieve parity across different social groups but leave aside individual facts, resulting in modifi…

Fairnessknowledge editingLanguage ModelingLanguage Modelling+2

Do Students Debias Like Teachers? On the Distillability of Bias Mitigation Methods

2025-10-30 · Jiali Cheng, Chirag Agarwal, Hadi Amiri arxiv

Knowledge distillation (KD) is an effective method for model compression and transferring knowledge between models. However, its effect on model's robustness against spurious correlations that degrade performance on out-…

Natural Language InferenceKnowledge DistillationImage ClassificationModel Compression

Bias Vector: Mitigating Biases in Language Models with Task Arithmetic Approach

2024-12-16 · Daiki Shirafuji, Makoto Takenaka, Shinya Taguchi

The use of language models (LMs) has increased considerably in recent years, and the biases and stereotypes in training data that are reflected in the LM outputs are causing social problems. In this paper, inspired by th…

Language ModelingLanguage ModellingMasked Language ModelingTask Arithmetic

Bias Mitigation or Cultural Commonsense? Evaluating LLMs with a Japanese Dataset

2025-09-29 · Taisei Yamamoto, Ryoma Kumon, Danushka Bollegala, Hitomi Yanaka arxiv

Large language models (LLMs) exhibit social biases, prompting the development of various debiasing methods. However, debiasing methods may degrade the capabilities of LLMs. Previous research has evaluated the impact of b…