Sociodemographic bias in language models (LMs) has the potential for harm when deployed in real-world settings. This paper presents a comprehensive survey of the past decade of research on sociodemographic bias in LMs, organized into a typology that facilitates examining the different aims: types of bias, quantifying bias, and debiasing techniques. We track the evolution of the latter two questions, then identify current trends and their limitations, as well as emerging techniques. To guide future research towards more effective and reliable solutions, and to help authors situate their work within this broad landscape, we conclude with a checklist of open questions.
One Persona, Many Cues, Different Results: How Sociodemographic Cues Impact LLM Personalization
Institute for Natural Language Processing, University of Stuttgart(斯图加特大学自然语言处理研究所);Institute for Logic, Language and Computation, University of Amsterdam(阿姆斯特丹大学逻辑、语言和计算研究所);Weizenbaum Institute(Weizenbaum研究所;慕尼黑机器学习中心;慕尼黑技术大学);Munich Center for Machine Learning;TU Munich
Bias Attribution in Filipino Language Models: Extending a Bias Interpretability Metric for Application on Agglutinative Languages
School of Computer Science, University of Birmingham(英国伯明翰大学计算机科学学院);Department of Information Systems and Computer Science, Ateneo de Manila University(马尼拉大学信息系统与计算机科学系)
Beyond Third-Person Audits: Situated Interaction Auditing for User-Centered LLM Bias Research
Department of Computer Science University of Chile Center for Artificial Intelligence (CENIA) Santiago, Chile(计算机科学系智利大学人工智能中心(CENIA)圣地亚哥,智利);Center for Artificial Intelligence (CENIA) Santiago, Chile(人工智能中心(CENIA)圣地亚哥,智利);Institute of Sociology Pontificia Universidad Católica de Chile Santiago, Chile(社会学研究所智利天主教大学圣地亚哥,智利)
Language (Technology) is Power: A Critical Survey of "Bias" in NLP
College of Information and Computer Sciences(信息与计算机科学学院);University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校);Microsoft Research(微软研究院);Cornell University(康奈尔大学);University of Maryland(马里兰大学)