通过增强偏见知识神经元，让大模型更公正而不失能力。
Knowing Bias, Doing Better: Mitigating Social Bias in LLMs via Know-Bias Neuron Enhancement
- 不压制偏见神经元，而是选择性增强它们来减少偏见。
- 仅需少量是非问题即可实现强去偏效果，无需重训练。
- 适用多种偏见类型和人群，保持模型整体性能不下降。
大型语言模型（LLMs）存在社会偏见，强化有害刻板印象，限制其安全部署。现有去偏方法多采用抑制策略，通过修改参数、提示或相关神经元来缓解偏见，但常导致鲁棒性差、泛化能力弱、数据效率低，并损害模型通用能力。本文提出轻量级且概念新颖的KnowBias框架，通过增强而非抑制编码偏见知识的神经元来缓解偏见。KnowBias利用少量偏见知识问答，结合归因分析识别出偏见神经元，并在推理时选择性增强这些神经元。该设计实现了强去偏效果，同时保留通用能力，跨偏见类型与人口群体具有良好泛化性，且高度数据高效——仅需数个简单的是非问题，无需重新训练。在多个基准测试和模型上实验表明，KnowBias在去偏性能上达到当前最优水平，且性能损失极小。代码与数据已开源。
原文摘要 · Abstract (English)
Large language models (LLMs) exhibit social biases that reinforce harmful stereotypes, limiting their safe deployment. Most existing debiasing methods adopt a suppressive paradigm by modifying parameters, prompts, or neurons associated with biased behavior; however, such approaches are often brittle, weakly generalizable, data-inefficient, and prone to degrading general capability. We propose \textbf{KnowBias}, a lightweight and conceptually distinct framework that mitigates bias by strengthening, rather than suppressing, neurons encoding bias-knowledge. KnowBias identifies neurons encoding bias knowledge using a small set of bias-knowledge questions via attribution-based analysis, and selectively enhances them at inference time. This design enables strong debiasing while preserving general capabilities, generalizes across bias types and demographics, and is highly data efficient, requiring only a handful of simple yes/no questions and no retraining. Experiments across multiple benchmarks and LLMs demonstrate consistent state-of-the-art debiasing performance with minimal utility degradation. Data and code are available at https://github.com/JP-25/KnowBias.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。