不同语言提示会影响大模型对心理健康评估的判断，中文更易引发偏见。
Language Shapes Mental Health Evaluations in Large Language Models
- 用中英文对比测试大模型心理评估倾向，发现语言影响判断标准。
- 中文提示下模型对污名化内容敏感度降低，抑郁判断更保守且常低估。
- 提醒开发者需关注多语言模型在敏感场景中的公平性与一致性。
多语言大语言模型（LLMs）在心理健康支持聊天机器人、筛查和内容审核等社会敏感领域应用日益广泛。这引发了一个可靠性问题：语义相同的心理健康输入在不同语言中是否会产生一致的评估结果，还是受语言相关社会文化背景影响而出现系统性偏差？本研究以英语-中文为对比，使用 GPT-4o 与 Qwen3-32B，在两级框架下进行分析：一是构建层面的评价取向，通过心理测量学污名量表衡量；二是决策层面的行为表现，包括二分类污名检测和四分类抑郁严重程度分类。结果显示，无论模型或量表，中文提示均引发更高的污名相关评分。在决策层面，中文提示降低了对污名内容的敏感性，并导致更保守的抑郁严重程度判断，产生更多低估错误。这些发现表明，提示语言可显著影响基于 LLM 的心理健康评估中的评价取向与下游行为。研究强调，评估多语言 LLM 时，不仅要看整体性能，更应检验其在社会敏感领域跨语言是否保持一致的评价标准。
原文摘要 · Abstract (English)
Multilingual large language models (LLMs) are increasingly used in socially sensitive mental health contexts, including support chatbots, screening, and content moderation. This raises a reliability question: do semantically equivalent mental health inputs elicit comparable evaluations across languages, or systematic shifts consistent with language-associated social and cultural contexts? We examine this question in an English-Chinese setting with GPT-4o and Qwen3-32B using a two-level framework: construct-level evaluative orientation, measured by psychometric stigma instruments, and decision-level behavior, measured by binary stigma detection and four-class depression severity classification. Across instruments and models, Chinese prompts elicit higher stigma-related scores than English prompts. At the decision level, Chinese prompts reduce sensitivity to stigmatizing content and produce more conservative depression severity judgments, leading to more under-estimation errors. These findings show that prompt language can shift both evaluative orientation and downstream behavior in LLM-based mental health evaluation. They highlight the need to evaluate multilingual LLMs not only for aggregate performance, but also for whether they apply comparable evaluative standards across languages in socially sensitive domains.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。