arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2210.14552cs.CL
收藏

一种基于刻板印象内容模型的稳健偏差消除方法

A Robust Bias Mitigation Procedure Based on the Stereotype Content Model

School of Informatics(信息学院);University of Edinburgh(爱丁堡大学)

Eddie L. Ungless, Amy Rafferty, Hrichika Nag, Björn Ross

摘要:

刻板印象内容模型(SCM)指出,我们倾向于将少数群体视为冷漠,无能或两者兼而有之。在本文中,我们适应现有的工作来证明,刻板印象的内容模型举行的上下文词嵌入,然后使用这些结果来评估微调的过程中,旨在推动语言模型远离刻板印象的少数群体。我们发现,SCM条款能够更好地捕捉偏见比人口统计不可知论的条款有关的愉快。此外,我们能够通过一个简单的微调过程来减少模型中存在的定型,这需要最少的人力和计算机资源,而不会损害下游性能。我们提出这项工作作为一个原型的去偏置程序,旨在消除需要先验知识的具体偏见的模型。

英文摘要:

The Stereotype Content model (SCM) states that we tend to perceive minority groups as cold, incompetent or both. In this paper we adapt existing work to demonstrate that the Stereotype Content model holds for contextualised word embeddings, then use these results to evaluate a fine-tuning process designed to drive a language model away from stereotyped portrayals of minority groups. We find the SCM terms are better able to capture bias than demographic agnostic terms related to pleasantness. Further, we were able to reduce the presence of stereotypes in the model through a simple fine-tuning procedure that required minimal human and computer resources, without harming downstream performance. We present this work as a prototype of a debiasing procedure that aims to remove the need for a priori knowledge of the specifics of bias in the model.

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。

↑