arXiv cs.CL· Sneha Oram, Ojaswita Bhushan, Tejomay Kishor Padole, Suyash P. Awate, Pushpak Bhattacharyya·· 2 天前AI 评分22
大语言模型在情感对话语境中的情感一致性分析
Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts
AI 导读
研究将 LLM 生成的文本作为查询回传给同一模型，用三种强度递增的虚假预设查询，在极端与中等情感两个维度上评估 Claude-3.5-haiku、GPT4o-mini 和 Mistral-7B 的自我一致性。结果显示三个模型表现均低于平均水平，易受查询中错误信念影响，中等情感内容下尤为明显；基于注意力分数的分析还显示模型优先级从评估转向生成。
正文 · 原文
Abstract:In this work, we conduct an analysis to examine the consistency of Large Language Models (LLMs) with respect to their own generated responses in an emotionally-driven conversational context. Specifically, the text generated by LLM is framed as a query to the same model, and its responses are subsequently assessed. This is performed with three queries across two dimensions of extreme and moderate emotions. The three queries are, in particular, false claim queries that contain inherently wrong assumptions (false presuppositions) in increasing order of intensity. Two commercial models, Claude-3.5-haiku, GPT4o-mini, and a medium-sized model, Mistral-7B, are considered in the study. Our findings indicate that LLMs exhibit below-average performance and remain vulnerable to false beliefs embedded within queries. This susceptibility is especially pronounced for moderate emotional content. Furthermore, an extended attention-score-based analysis highlights a shift in models' priority from evaluative to generative. The results raise important considerations for LLMs' deployment in high-stakes, emotionally sensitive contexts.
| Comments: | Accepted in MultiPsyche Workshop (In Conjunction with AACL-IJCNLP 2026) |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2605.06476 [cs.CL] |
| (or arXiv:2605.06476v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.06476 arXiv-issued DOI via DataCite |
Submission history
From: Sneha Oram [view email]
[v1]
Thu, 7 May 2026 16:01:48 UTC (1,242 KB)
[v2]
Thu, 15 Oct 2026 02:23:48 UTC (1,884 KB)
来源：arXiv cs.CL · arxiv.org