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 2605.05403cs.AI
收藏

当有益变成谄媚:谄媚是大型语言模型中社会契合与认知完整性之间的边界失败

When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models

Duke University(杜克大学)

Jiechen Li, Catherine A. Barry, Rishika Randev, Janet Chen, Ella Jorgensen, Brinnae Bent

AI总结:

本文探讨大型语言模型中谄媚现象作为社会契合与认知完整性之间边界失败的问题,提出三条件框架以界定谄媚,并讨论其分类、评估及缓解策略。

摘要:

本文主张大型语言模型中的谄媚是社会契合与认知完整性之间的边界失败。现有研究通常通过外部行为如同意错误用户信念、立场反转或偏离客观正确性标准来定义谄媚。这些定义仅捕捉到谄媚的显性形式,而未明确涉及认知完整性和社会契合的更微妙边界失败。我们主张谄媚不应仅理解为同意,而应视为一种使独立认知判断失衡的契合行为。为明确此边界,我们提出三条件框架。首先,用户以信念、偏好或自我概念的形式表达提示。其次,模型通过契合行为向该提示偏移。第三,这种偏移损害认知准确性、独立推理或适当纠正。我们还引入一种分类体系,用于对谄媚进行分类,包括契合目标、机制和严重性。本文最后讨论了对契合评估的影响,并主张边界意识评估、结构化评分表和缓解策略,同时将这些提议置于其他谄媚观点的背景下。

英文摘要:

This position paper argues that sycophancy in LLMs is a boundary failure between social alignment and epistemic integrity. Existing work often operationalizes sycophancy through external behavior such as agreement with incorrect user beliefs, position reversals, or deviation from an objective standard of correctness. These formulations capture overt forms of the phenomenon but leave subtler boundary failures involving epistemic integrity and social alignment underspecified. We argue that sycophancy should be defined not by agreement or factual incorrectness alone, but by cue-driven degradation of epistemic integrity. To clarify this boundary, we propose a three-condition decision rule requiring (1) a user cue, (2) a model response that shifts toward that cue, and (3) a resulting degradation of epistemic integrity. Building on this definition, we also introduce a taxonomy consisting of alignment targets, mechanisms, and severity to characterize cases identified as sycophantic. An initial 48-item annotation pilot provides preliminary evidence that the framework can be operationalized using structured guidelines, although broader empirical validation remains necessary. The paper concludes by discussing implications for alignment evaluation, including boundary-aware assessment, structured rubrics, and mitigation strategies, while situating these proposals alongside alternative views of sycophancy.

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。

↑