Multi-Dimensional Gender Bias Classification
Emily Dinan∗, Angela Fan †, Ledell Wu, Jason Weston, Douwe Kiela, Adina Williams Facebook AI Research †Laboratoire Lorrain d’Informatique et Applications (LORIA) Joint first authors.
Abstract
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.
中文速览
NLP模型在学习文本数据时会不自觉地习得性别偏见，但现有研究往往把"性别偏见"当作一个笼统的概念来处理，缺乏细粒度的区分。这项工作提出了一套三维框架，将文本中的性别偏见拆解为"被谈论者的性别"、"被说话对象的性别"和"说话者自身的性别"三个独立维度，并据此对八个大规模数据集进行自动标注，同时众包收集了一个新的基准测试集（MDGender），包含从不同性别视角改写的句子。基于这套框架训练出的多维度性别偏见分类器，被验证可有效应用于控制生成模型的性别倾向、检测任意文本中的性别偏见，以及分析冒犯性语言与性别之间的关联。这项工作的重要性在于，它为性别偏见的精细化识别与缓解提供了可复用的工具和公开数据集，有助于推动NLP系统朝着更公平、更包容的方向发展。
原文 arXiv:2005.00614;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2005.00614v1