arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.12428cs.CLcs.AI
收藏

大型语言模型中的从众效应

Conformity in Large Language Models

University of Cambridge(剑桥大学);University of Sheffield(谢菲尔德大学)

Xiaochen Zhu, Caiqi Zhang, Tom Stafford, Nigel Collier, Andreas Vlachos

AI总结:

本研究通过改编心理学实验,发现大型语言模型普遍存在从众效应,且不确定性越高越易从众;指令调整可降低从众,而自然语调会加剧从众,并提出两种干预措施以缓解该问题。

摘要:

从众效应描述了个体倾向于使其回答与多数人保持一致的现象。研究大型语言模型(LLM)中的这种偏差至关重要,因为LLM越来越多地被用作对话伙伴,在各种信息检索和决策任务中提高生产力。因此,对错误回答的从众可能会损害其有效性。在本文中,我们改编心理学实验来考察流行LLM中的从众程度。我们的研究结果显示,所有测试的模型在不同知识领域中都表现出不同程度的对多数意见的从众,无论其初始选择或正确性如何。值得注意的是,我们首次证明,当LLM对其自身预测更加不确定时,它们更倾向于从众。我们进一步探讨了影响从众的因素,如训练范式和输入特征,发现指令调整的模型对从众的敏感性较低,而增加多数意见语调的自然性会加剧从众。最后,我们提出了两种干预措施,“魔鬼代言人”和“问题蒸馏”,以减轻从众效应,为构建更稳健的语言模型提供见解。

英文摘要:

The conformity effect describes the tendency of individuals to align their responses with the majority. Studying this bias in large language models (LLMs) is crucial, as LLMs are increasingly used in various information-seeking and decision-making tasks as conversation partners to improve productivity. Thus, conformity to incorrect responses can compromise their effectiveness. In this paper, we adapt psychological experiments to examine the extent of conformity in popular LLMs. Our findings reveal that all tested models exhibit varying levels of conformity toward the majority, regardless of their initial choice or correctness, across different knowledge domains. Notably, we are the first to show that LLMs are more likely to conform when they are more uncertain in their own prediction. We further explore factors that influence conformity, such as training paradigms and input characteristics, finding that instruction-tuned models are less susceptible to conformity, while increasing the naturalness of majority tones amplifies conformity. Finally, we propose two interventions, Devil's Advocate and Question Distillation, to mitigate conformity, providing insights into building more robust language models.

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。

↑