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 2602.06260cs.CLcs.AI
收藏

单方面论点能否引导大语言模型产生回应变化?

Can One-sided Arguments Lead to Response Change in Large Language Models?

VMware Research(VMware研究)

Pedro Cisneros-Velarde

AI总结:

本文研究了通过单方面论点是否能引导大语言模型改变其回应观点,发现不同模型和主题下均存在观点引导现象,而转向其他论点可减少这种影响。

摘要:

辩论性问题需要多个观点来表达平衡的回答。大语言模型(LLMs)可以提供平衡的回答,但也可以采取单一一致的观点或拒绝回答。在本文中,我们研究是否可以通过简单直观的方式引导初始回应转向特定观点:仅提供支持该观点的一方论点。我们的系统研究有三个维度:(i)在LLM回应中诱导出哪种立场,(ii)辩论性问题是如何表述的,(iii)论点是如何展示的。我们构建了一个小数据集,并显著发现,对于不同的模型、论点数量和主题,(i)-(iii)中均会发生观点引导。转向其他论点一致地减少了观点引导。

英文摘要:

Polemic questions need more than one viewpoint to express a balanced answer. Large Language Models (LLMs) can provide a balanced answer, but also take a single aligned viewpoint or refuse to answer. In this paper, we study if such initial responses can be steered to a specific viewpoint in a simple and intuitive way: by only providing one-sided arguments supporting the viewpoint. Our systematic study has three dimensions: (i) which stance is induced in the LLM response, (ii) how the polemic question is formulated, (iii) how the arguments are shown. We construct a small dataset and remarkably find that opinion steering occurs across (i)-(iii) for diverse models, number of arguments, and topics. Switching to other arguments consistently decreases opinion steering.

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。

↑