Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods
Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods
复制标题
DOI：
10.18653/v1/n18-2003
复制
发表时间：
2018-04
期刊：
影响因子：
--
通讯作者：
Jieyu Zhao;Tianlu Wang;Mark Yatskar;Vicente Ordonez;Kai-Wei Chang
中科院分区：
文献类型：
--
作者：
Jieyu Zhao;Tianlu Wang;Mark Yatskar;Vicente Ordonez;Kai-Wei Chang
在本文中，我们介绍了一个新的基准，以共同参考决议，重点是性别偏见，Winobias。我们的语料库包含Winograd-Schema风格的句子，该句子的实体与他们的职业相对应（例如护士，医生，木匠）。我们证明，基于规则的，功能丰富的和神经核心系统的所有链接性别代词与抗疾病型实体的准确性更高的亲构型实体的链接，平均差异为F1分数21.1。最后，我们展示了一种数据启发方法，该方法与现有的单词插入式伪造技术相结合，消除了这些系统在Winobias中所证明的偏差，而不会显着影响其在现有数据集上的性能。
In this paper, we introduce a new benchmark for co-reference resolution focused on gender bias, WinoBias. Our corpus contains Winograd-schema style sentences with entities corresponding to people referred by their occupation (e.g. the nurse, the doctor, the carpenter). We demonstrate that a rule-based, a feature-rich, and a neural coreference system all link gendered pronouns to pro-stereotypical entities with higher accuracy than anti-stereotypical entities, by an average difference of 21.1 in F1 score. Finally, we demonstrate a data-augmentation approach that, in combination with existing word-embedding debiasing techniques, removes the bias demonstrated by these systems in WinoBias without significantly affecting their performance on existing datasets.