arXiv 2508.061962026-06-15cs.CLcs.HC

EiCAP：超越流畅性，通过心理学基础的多轮对话探究和提升大语言模型的情感智能

EiCAP: Beyond Fluency, Probing and Improving Emotional Intelligence in LLMs via Psychologically Grounded Multi-Turn Dialogue

Qatar Computing Research Institute (QCRI), Hamad Bin Khalifa University（卡塔尔计算研究所（QCRI），哈马德·本·卡伊夫大学）；University of Illinois Urbana-Champaign (UIUC)（伊利诺伊大学厄巴纳-香槟分校）

Nizi Nazar, Pardis Sadat Zahraei, Dilek Hakkani-Tür, Natasa Milic-Frayling, Ehsaneddin Asgari

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出基于心理学六层情感智能分类法的EiCAP框架，包含评估基准EiCAP-Bench和微调语料EiCAP-SFT，发现通用对话微调不提升情感智能，而基于情感智能的LoRA微调显著提升模型在所有24个子类别上的表现。

摘要：

大语言模型越来越多地用于情感敏感角色，包括心理健康支持、教育和危机响应，但它们缺乏评估或提升情感智能（EI）的原则性框架。我们引入EiCAP，一个统一的、基于心理学的六层EI分类法，并实现为两个互补资源。EiCAP-Bench是一个多轮、一对三强制选择评估套件，包含3,174个探针，涵盖24个子类别和跨轮依赖关系，反映真实对话EI需求。EiCAP-SFT是一个152,820轮对话的监督语料库，与同一分类法对齐，支持可控、可解释的微调。两个关键发现出现。首先，通用对话监督微调不会赋予EI：在UltraChat上微调在24个子类别中均无显著提升，宏观得分为24.6%，接近随机水平25%。其次，直接对Qwen-2.5-7B-Base应用基于EI的LoRA（使用约0.8%的参数）在全部24个子类别上取得显著提升，宏观得分达到75.33%，比Base提升51.7个百分点，比Instruct提升37.1个百分点。关键的是，消融实验表明UltraChat预训练阶段适得其反，使性能降低21.4个百分点：直接基于EI的训练既必要又充分。

英文摘要：

Large Language Models increasingly serve in emotionally sensitive roles, including mental health support, education, and crisis response, yet they lack a principled framework for assessing or improving Emotional Intelligence (EI). We introduce EiCAP, a unified, psychologically grounded six-layer EI taxonomy operationalized into two complementary resources. EiCAP-Bench is a multi-turn, one-vs-three forced-choice evaluation suite with 3,174 probes across 24 subcategories and cross-turn dependencies that reflect real conversational EI demands. EiCAP-SFT is a 152,820-dialogue supervision corpus aligned to the same taxonomy, enabling controlled, interpretable fine-tuning. Two key findings emerge. First, generic conversational supervised fine-tuning does not confer EI: fine-tuning on UltraChat yields no significant gain in any of the 24 subcategories, with a macro score of 24.6%, near the chance level of 25%. Second, applying EI-grounded LoRA, using approximately 0.8% of parameters, directly to Qwen-2.5-7B-Base achieves significant gains in all 24 subcategories, reaching a macro score of 75.33%, a gain of 51.7 percentage points over Base and 37.1 percentage points over Instruct. Crucially, an ablation shows that the UltraChat pre-stage is counterproductive, reducing performance by 21.4 percentage points: direct EI-grounded training is both necessary and sufficient.

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。