标记人物角色：使用自然语言提示测量语言模型中的刻板印象
Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models
AI总结：
本文提出Marked Personas方法，通过提示LLM生成标记与无标记群体的人物角色并比较差异词，无需词典或标注即可测量交叉群体的刻板印象，发现GPT模型比人类生成更多种族刻板印象，并揭示他者化、异域化等表征性危害。
摘要：
为了识别和减轻大型语言模型（LLMs）带来的危害，我们需要理解LLM输出中刻板印象的普遍性和细微差别。为此，我们提出了Marked Personas，这是一种基于提示的方法，用于在没有任何词典或数据标注的情况下，测量LLM中针对交叉人口群体的刻板印象。该方法以社会语言学中的标记性（markedness）概念为基础（该概念刻画了显性语言标记类别与无标记默认类别之间的区别），我们提出的方法包含两个步骤：1）提示LLM生成目标人口群体的人物角色（personas），即自然语言描述，同时生成无标记默认群体的人物角色；2）识别能够显著区分目标群体人物角色与相应无标记群体人物角色的词语。我们发现，与使用相同提示的人工撰写描述相比，GPT-3.5和GPT-4生成的描述包含更高比例的种族刻板印象。区分标记群体（非白人、非男性）人物角色的词语反映了将这些人口群体他者化（othering）和异域化（exoticizing）的模式。交叉性视角进一步揭示了主导边缘化群体描述的刻板套路，例如热带主义（tropicalism）和少数族裔女性的过度性化（hypersexualization）。这些表征性危害对故事生成等下游应用具有令人担忧的影响。
英文摘要：
To recognize and mitigate harms from large language models (LLMs), we need to understand the prevalence and nuances of stereotypes in LLM outputs. Toward this end, we present Marked Personas, a prompt-based method to measure stereotypes in LLMs for intersectional demographic groups without any lexicon or data labeling. Grounded in the sociolinguistic concept of markedness (which characterizes explicitly linguistically marked categories versus unmarked defaults), our proposed method is twofold: 1) prompting an LLM to generate personas, i.e., natural language descriptions, of the target demographic group alongside personas of unmarked, default groups; 2) identifying the words that significantly distinguish personas of the target group from corresponding unmarked ones. We find that the portrayals generated by GPT-3.5 and GPT-4 contain higher rates of racial stereotypes than human-written portrayals using the same prompts. The words distinguishing personas of marked (non-white, non-male) groups reflect patterns of othering and exoticizing these demographics. An intersectional lens further reveals tropes that dominate portrayals of marginalized groups, such as tropicalism and the hypersexualization of minoritized women. These representational harms have concerning implications for downstream applications like story generation.
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。
