通过调整注意力分布，让大模型在偏见和无偏场景下表现一致。
KLAAD: Refining Attention Mechanisms to Reduce Societal Bias in Generative Language Models
- 基于注意力机制设计无权重修改的去偏框架，对齐刻板与反刻板句式注意力
- 在BBQ和BOLD基准上显著降低社会偏见，语言质量几乎不受影响
- 适合关注生成模型公平性、不想改模型结构的研究者
大型语言模型在输出中常表现出社会偏见，引发公平性与潜在伤害的伦理担忧。本文提出一种基于注意力的去偏框架KLAAD（KL-Attention Alignment Debiasing），无需直接修改模型权重，即可隐式对齐刻板与反刻板句子对的注意力分布。KLAAD采用交叉熵、KL散度与三元组损失组成的复合训练目标，引导模型在有偏与无偏语境中保持一致的注意力模式，同时维持语言流畅性与连贯性。实验表明，KLAAD在BBQ和BOLD基准上均实现更好的去偏效果，且对语言建模质量影响极小。结果表明，注意力层面的对齐提供了一种原则性方法，用于缓解生成式语言模型中的偏见问题。
原文摘要 · Abstract (English)
Large language models (LLMs) often exhibit societal biases in their outputs, prompting ethical concerns regarding fairness and harm. In this work, we propose KLAAD (KL-Attention Alignment Debiasing), an attention-based debiasing framework that implicitly aligns attention distributions between stereotypical and anti-stereotypical sentence pairs without directly modifying model weights. KLAAD introduces a composite training objective combining Cross-Entropy, KL divergence, and Triplet losses, guiding the model to consistently attend across biased and unbiased contexts while preserving fluency and coherence. Experimental evaluation of KLAAD demonstrates improved bias mitigation on both the BBQ and BOLD benchmarks, with minimal impact on language modeling quality. The results indicate that attention-level alignment offers a principled solution for mitigating bias in generative language models.
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。