多语言语言模型中的社会偏见:综述
Social Bias in Multilingual Language Models: A Survey
AI总结:
本综述系统梳理多语言模型社会偏见评估与缓解的相关研究,指出当前方法在语言偏好、实验匮乏等方面的缺口,提出未来需提升研究包容性与跨文化适配性的方向。
摘要:
预训练多语言模型表现出与处理英文文本的模型相同的社会偏见。本系统综述分析了将偏见评估与缓解方法拓展至多语言及非英语语境的新兴研究。我们从语言多样性、文化意识,以及评估指标与缓解技术的选择维度对这些研究展开考察。本综述揭示了该领域主流方法设计选择中的缺口(例如对特定语言的偏好、多语言缓解实验的匮乏),同时梳理了在跨语言与跨文化适配偏见基准过程中遇到的常见问题及已采用的解决方案。基于研究发现的启示,我们规划了未来研究方向,以增强多语言偏见研究的包容性、跨文化适配性,并与最前沿的NLP进展保持对齐。
英文摘要:
Pretrained multilingual models exhibit the same social bias as models processing English texts. This systematic review analyzes emerging research that extends bias evaluation and mitigation approaches into multilingual and non-English contexts. We examine these studies with respect to linguistic diversity, cultural awareness, and their choice of evaluation metrics and mitigation techniques. Our survey illuminates gaps in the field's dominant methodological design choices (e.g., preference for certain languages, scarcity of multilingual mitigation experiments) while cataloging common issues encountered and solutions implemented in adapting bias benchmarks across languages and cultures. Drawing from the implications of our findings, we chart directions for future research that can reinforce the multilingual bias literature's inclusivity, cross-cultural appropriateness, and alignment with state-of-the-art NLP advancements.
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。