站在他人的角度：视角采择如何引导大型语言模型减少毒性及偏见
Walking in Others' Shoes: How Perspective-Taking Guides Large Language Models in Reducing Toxicity and Bias
AI总结：
针对LLMs生成内容的毒性与偏见问题，提出视角采择提示（PeT）策略，通过整合人类视角自我调节，在商业及开源模型上显著降低毒性（达89%）与偏见（达73%），优于五个强基线。
摘要：
大型语言模型（LLMs）生成内容中常见的毒性和社会偏见，亟需减少其危害的策略。现有解决方案通常要求对模型进行白盒访问或进行大量训练，这对于前沿商业LLMs而言并不切实际。此外，现有的提示方法依赖于外部工具反馈，且无法同时减轻毒性和偏见。受社会心理学原理启发，我们提出了一种名为“视角采择提示（PeT）”的新策略，该策略激励LLMs整合多样的人类视角并自我调节其回应。这种自我纠正机制可显著降低LLMs回应中的毒性（最高达89%）和偏见（最高达73%）。我们在两个商业LLMs（ChatGPT和GLM）及三个开源LLMs上进行了严格的评估和消融研究，结果显示PeT在生成较少有害回应方面优于五个强基线，展现出其优越性。
英文摘要：
The common toxicity and societal bias in contents generated by large language models (LLMs) necessitate strategies to reduce harm. Present solutions often demand white-box access to the model or substantial training, which is impractical for cutting-edge commercial LLMs. Moreover, prevailing prompting methods depend on external tool feedback and fail to simultaneously lessen toxicity and bias. Motivated by social psychology principles, we propose a novel strategy named \textbf{perspective-taking prompting (\textsc{PeT})} that inspires LLMs to integrate diverse human perspectives and self-regulate their responses. This self-correction mechanism can significantly diminish toxicity (up to $89\%$) and bias (up to $73\%$) in LLMs' responses. Rigorous evaluations and ablation studies are conducted on two commercial LLMs (ChatGPT and GLM) and three open-source LLMs, revealing \textsc{PeT}'s superiority in producing less harmful responses, outperforming five strong baselines.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
