arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.18403cs.CY
收藏

去WEIRD化LLM实体推荐

UnWEIRDing LLM Entity Recommendations

Microsoft(微软)

Aayush Kumar, Sanket Mhatre

AI总结:

本文研究了LLM在现实世界实体推荐中的文化偏见,通过WEIRD框架评估并应用多元提示策略以减轻偏见,发现不同模型的偏见减轻效果不一致。

摘要:

大型语言模型已被广泛用于句子完成等写作任务。尽管这可以提高写作效率,但先前研究显示,LLM生成的建议可能表现出文化偏见,这可能使用户难以察觉,尤其是在非英语母语者教育情境中。虽然先前工作已研究了LLM道德价值观对齐中的偏见,但我们旨在调查LLM在现实世界实体推荐中的文化偏见。为此,我们使用WEIRD(西方、受过教育、工业化、富裕和民主)框架,评估各种LLM在细粒度实体数据集上的推荐情况,并应用多元提示策略来减轻这些偏见。我们的结果表明,尽管此类提示策略能减少此类偏见,但这种减少在不同模型之间并不一致,某些类型的实体推荐比其他类型更具有偏见。

英文摘要:

Large Language Models have been widely been adopted by users for writing tasks such as sentence completions. While this can improve writing efficiency, prior research shows that LLM-generated suggestions may exhibit cultural biases which may be difficult for users to detect, especially in educational contexts for non-native English speakers. While such prior work has studied the biases in LLM moral value alignment, we aim to investigate cultural biases in LLM recommendations for real-world entities. To do so, we use the WEIRD (Western, Educated, Industrialized, Rich and Democratic) framework to evaluate recommendations by various LLMs across a dataset of fine-grained entities, and apply pluralistic prompt-based strategies to mitigate these biases. Our results indicate that while such prompting strategies do reduce such biases, this reduction is not consistent across different models, and recommendations for some types of entities are more biased than others.

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。

↑