arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.00614cs.CL
收藏

多维性别偏见分类

Multi-Dimensional Gender Bias Classification

Facebook AI Research(Facebook人工智能研究院);Laboratoire Lorrain d’Informatique et Applications (LORIA)(洛林计算机科学及其应用实验室(LORIA))

Emily Dinan, Angela Fan, Ledell Wu, Jason Weston, Douwe Kiela, Adina Williams

摘要:

机器学习模型被训练来寻找数据中的模式。NLP模型在训练有性别偏见的文本时,可能会无意中学习到社会上不受欢迎的模式。在这项工作中,我们提出了一个总体框架,分解文本中的性别偏见沿几个语用和语义方面:从性别的人被谈论的偏见,从性别的人被说话的偏见,从性别的扬声器的偏见。使用这个细粒度的框架,我们自动注释八个大规模的数据集与性别信息。此外,我们还收集了一个新颖的、众包的话语级性别重写评估基准。区分多个维度的性别偏见是很重要的,因为它使我们能够训练更细粒度的性别偏见分类器。我们证明了我们的分类器对于各种重要的应用都是有价值的,例如控制生成模型中的性别偏见,检测任意文本中的性别偏见,并在性别方面揭示攻击性语言。

英文摘要:

Machine learning models are trained to find patterns in data. NLP models can inadvertently learn socially undesirable patterns when training on gender biased text. In this work, we propose a general framework that decomposes gender bias in text along several pragmatic and semantic dimensions: bias from the gender of the person being spoken about, bias from the gender of the person being spoken to, and bias from the gender of the speaker. Using this fine-grained framework, we automatically annotate eight large scale datasets with gender information. In addition, we collect a novel, crowdsourced evaluation benchmark of utterance-level gender rewrites. Distinguishing between gender bias along multiple dimensions is important, as it enables us to train finer-grained gender bias classifiers. We show our classifiers prove valuable for a variety of important applications, such as controlling for gender bias in generative models, detecting gender bias in arbitrary text, and shed light on offensive language in terms of genderedness.

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。

↑