arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.14727cs.CYcs.CLcs.LG
收藏

大型语言模型中的受保护群体偏见与刻板印象

Protected group bias and stereotypes in Large Language Models

Apple(苹果公司);MIT(麻省理工学院);University of Manchester(曼彻斯特大学)

Hadas Kotek, David Q. Sun, Zidi Xiu, Margit Bowler, Christopher Klein

AI总结:

本文研究了大型语言模型在伦理和公平性领域的受保护群体偏见,通过收集并人工标注超过1万个句子补全,发现模型不仅反映且放大了社会偏见,并在回复少数群体查询时过度谨慎,强调多样性而掩盖其他特征。

摘要:

随着现代大型语言模型(LLMs)在多个领域打破了许多最先进的基准测试,本文研究了它们在伦理和公平性领域的行为,重点关注受保护群体偏见。我们进行了一项分为两部分的研究:首先,我们请求模型续写描述来自不同受保护群体(包括性别、性取向、宗教和种族)的个体职业的句子。其次,我们让模型生成关于从事不同类型职业的个体的故事。我们收集了一个公开可用的LLM生成的超过1万个句子补全,并对其进行了人工标注。我们在少数群体中发现了偏见,尤其是在性别和性取向领域,以及模型生成内容中的西方偏见。该模型不仅反映了社会偏见,而且似乎放大了这些偏见。此外,模型在回复与少数群体相关的查询时过于谨慎,提供的回复强烈强调多样性和公平性,以至于其他群体特征被掩盖。这表明人为限制潜在有害输出本身可能导致伤害,应以谨慎和受控的方式应用。

英文摘要:

As modern Large Language Models (LLMs) shatter many state-of-the-art benchmarks in a variety of domains, this paper investigates their behavior in the domains of ethics and fairness, focusing on protected group bias. We conduct a two-part study: first, we solicit sentence continuations describing the occupations of individuals from different protected groups, including gender, sexuality, religion, and race. Second, we have the model generate stories about individuals who hold different types of occupations. We collect >10k sentence completions made by a publicly available LLM, which we subject to human annotation. We find bias across minoritized groups, but in particular in the domains of gender and sexuality, as well as Western bias, in model generations. The model not only reflects societal biases, but appears to amplify them. The model is additionally overly cautious in replies to queries relating to minoritized groups, providing responses that strongly emphasize diversity and equity to an extent that other group characteristics are overshadowed. This suggests that artificially constraining potentially harmful outputs may itself lead to harm, and should be applied in a careful and controlled manner.

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。

↑