评测大模型在个性化教育中的偏见，发现其教学内容对不同群体存在系统性偏差。
LLMs are Biased Teachers: Evaluating LLM Bias in Personalized Education
- 用两种新指标量化模型生成内容的偏见程度。
- 收入和残疾状态相关偏差最大，最高MAB达0.32，MDB达0.45。
- 即使顶尖模型也普遍存在偏见，适合教育科技开发者参考。
随着大语言模型在教育领域的广泛应用，其固有偏见问题日益受到关注。本文评估了大模型在个性化教育场景中作为‘教师’角色时的偏见表现，重点关注其针对不同人口统计群体（包括种族、族裔、性别、残疾状况、收入水平及国籍）生成和选择教育内容的方式。我们引入并应用了两种偏见评分指标——平均绝对偏见（MAB）与最大差异偏见（MDB），分析了9个主流开放与闭源大模型。实验基于超过17,000条跨难度与主题的教育解释内容，结果表明模型可能通过延续或反转有害刻板印象而损害学生学习效果。所有前沿模型均表现出显著偏见，其中收入相关偏见的MAB最高，而收入与残疾状态相关的MDB最高。在性别与种族/族裔方面，偏见最低。
原文摘要 · Abstract (English)
With the increasing adoption of large language models (LLMs) in education, concerns about inherent biases in these models have gained prominence. We evaluate LLMs for bias in the personalized educational setting, specifically focusing on the models' roles as "teachers." We reveal significant biases in how models generate and select educational content tailored to different demographic groups, including race, ethnicity, sex, gender, disability status, income, and national origin. We introduce and apply two bias score metrics--Mean Absolute Bias (MAB) and Maximum Difference Bias (MDB)--to analyze 9 open and closed state-of-the-art LLMs. Our experiments, which utilize over 17,000 educational explanations across multiple difficulty levels and topics, uncover that models potentially harm student learning by both perpetuating harmful stereotypes and reversing them. We find that bias is similar for all frontier models, with the highest MAB along income levels while MDB is highest relative to both income and disability status. For both metrics, we find the lowest bias exists for sex/gender and race/ethnicity.
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。