LLMs在性别与性少数群体描述中放大刻板印象，影响创作真实性。
LLMs Reproduce Stereotypes of Sexual and Gender Minorities
- 基于刻板印象内容模型，分析人类与LLM对多元性别群体的负面认知
- 在文本生成中，LLM表现出对性少数群体的刻板化描述，加剧代表性伤害
- 揭示了大模型在创意写作场景中的潜在偏见，适合关注伦理与公平的研究者
大量研究已发现自然语言处理系统存在显著性别偏见。现有研究多采用二元、本质主义的性别观：将性别简化为‘男性’和‘女性’两类，混淆性别与生理性别，忽视多元性身份。然而性别与性取向本就存在于光谱之中。本文基于广泛使用的社会心理学模型——刻板印象内容模型，研究大语言模型（LLMs）对非二元性别与性少数群体的偏见。我们发现，针对社会认知的英文调查问题会从人类和LLM中引发对性少数群体的更多负面刻板印象。进一步扩展至更真实的文本生成场景，分析显示LLM在此类任务中生成了对性少数群体的刻板化表征，表明其在创造性写作中会放大代表性伤害，而该场景正是当前许多大模型宣传的重点应用之一。
原文摘要 · Abstract (English)
A large body of research has found substantial gender bias in NLP systems. Most of this research takes a binary, essentialist view of gender: limiting its variation to the categories _men_ and _women_, conflating gender with sex, and ignoring different sexual identities. But gender and sexuality exist on a spectrum, so in this paper we study the biases of large language models (LLMs) towards sexual and gender minorities beyond binary categories. Grounding our study in a widely used social psychology model -- the Stereotype Content Model -- we demonstrate that English-language survey questions about social perceptions elicit more negative stereotypes of sexual and gender minorities from both humans and LLMs. We then extend this framework to a more realistic use case: text generation. Our analysis shows that LLMs generate stereotyped representations of sexual and gender minorities in this setting, showing that they amplify representational harms in creative writing, a widely advertised use for LLMs.
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。