研究GPT对国籍与精神障碍的交叉偏见，发现对北韩人尤其不公。
Assessing GPT's Bias Towards Stigmatized Social Groups: An Intersectional Case Study on Nationality Prejudice and Psychophobia
- 用结构化提示测试GPT对美朝人与心理障碍者的反应差异
- 北韩人+心理障碍者时共情度最低，偏差显著加剧
- 警示需关注模型对多重身份交织的敏感性
近期研究已分别揭示基础大语言模型（LLMs）对特定国籍及污名化社会群体存在显著偏见。本研究探讨了这些偏见在广泛使用的GPT-3.5/4/4o模型输出中的伦理影响。通过结构化提示序列，评估模型对涉及美国与朝鲜国籍且伴有不同心理障碍情境的响应。结果表明，对朝鲜人表现出更高的负面偏见，尤其当心理障碍因素同时存在时，共情水平显著降低。这凸显了在设计大语言模型时，需具备对交叉身份的精细理解能力。
原文摘要 · Abstract (English)
Recent studies have separately highlighted significant biases within foundational large language models (LLMs) against certain nationalities and stigmatized social groups. This research investigates the ethical implications of these biases intersecting with outputs of widely-used GPT-3.5/4/4o LLMS. Through structured prompt series, we evaluate model responses to several scenarios involving American and North Korean nationalities with various mental disabilities. Findings reveal significant discrepancies in empathy levels with North Koreans facing greater negative bias, particularly when mental disability is also a factor. This underscores the need for improvements in LLMs designed with a nuanced understanding of intersectional identity.
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。