Scholarlink

<p>We evaluated the ability of three leading LLMs (GPT-4o, Gemini 2.0 Experimental, and Claude 3.5 Sonnet) to recognize human facial expression using the NimStim dataset. GPT and Gemini matched or exceeded human performance, especially for calm/neutral and surprise. All models showed strong agreement with ground truth, though fear was often misclassified. Findings underscore the growing socioemotional competence of LLMs and their potential for healthcare applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Evaluating the performance of general purpose large language models in identifying human facial emotions

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月16日

查看全文
  • Benjamin W. Nelson,
  • Ari Winbush,
  • Steven Siddals,
  • Matthew Flathers,
  • Nicholas B. Allen,
  • John Torous

摘要

We evaluated the ability of three leading LLMs (GPT-4o, Gemini 2.0 Experimental, and Claude 3.5 Sonnet) to recognize human facial expression using the NimStim dataset. GPT and Gemini matched or exceeded human performance, especially for calm/neutral and surprise. All models showed strong agreement with ground truth, though fear was often misclassified. Findings underscore the growing socioemotional competence of LLMs and their potential for healthcare applications.