Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods

Published in NAACL, 2018

Recommended citation: Bibtex

Jieyu Zhao, Tianlu Wang, Mark Yatskar, Vicente Ordonez, Kai-Wei Chang


We introduce a new benchmark, WinoBias, for coreference resolution focused on gender bias. 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 coreference benchmark datasets. Our dataset and code are avialable at (The website is under construction now. Please refer to the github repo first.)


@inproceedings{BCWS18, author = {Jieyu Zhao and Tianlu Wang and Mark Yatskar and Vicente Ordonez and Kai-Wei Chang}, title = {Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods}, booktitle = {NAACL}, year = {2018}, }