Gender Bias in Coreference Resolution

Rachel Rudinger, Jason Naradowsky, Brian Leonard, Benjamin Van Durme
4/25/2018
cs.CL

Official code: rudinger/winogender-schemas (83 stars, MIT, last commit 2019-05-14).

Abstract

We present an empirical study of gender bias in coreference resolution systems. We first introduce a novel, Winograd schema-style set of minimal pair sentences that differ only by pronoun gender. With these "Winogender schemas," we evaluate and confirm systematic gender bias in three publicly-available coreference resolution systems, and correlate this bias with real-world and textual gender statistics.

View on arXivView PDF

Code Implementations(1)

rudinger/winogender-schemasOfficial100%
MIT

Data for evaluating gender bias in coreference resolution systems.

8314PythonApr 11, 20187 years agoMIT

Similar Papers

Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns

86% match2023

VisoGender: A dataset for benchmarking gender bias in image-text pronoun resolution

84% match1 repo132023

Gender Inflected or Bias Inflicted: On Using Grammatical Gender Cues for Bias Evaluation in Machine Translation

84% match2023

Voice, Bias, and Coreference: An Interpretability Study of Gender in Speech Translation

84% match2025

Gender-Inclusive Grammatical Error Correction through Augmentation

84% match1 repo82023

Gender Bias in Large Language Models across Multiple Languages

84% match2024

Cite this paper

@article{rudinger2018gender,
  title  = {Gender Bias in Coreference Resolution},
  author = {Rachel Rudinger and Jason Naradowsky and Brian Leonard and Benjamin Van Durme},
  year   = {2018},
  eprint = {1804.09301},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1804.09301}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1804.09301.svg)](https://codeofpaper.com/paper/1804.09301)

Discussion