English

Gender Bias in Coreference Resolution

Computation and Language 2018-04-26 v1

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.

Cite

@article{arxiv.1804.09301,
  title  = {Gender Bias in Coreference Resolution},
  author = {Rachel Rudinger and Jason Naradowsky and Brian Leonard and Benjamin Van Durme},
  journal= {arXiv preprint arXiv:1804.09301},
  year   = {2018}
}

Comments

Accepted to NAACL-HLT 2018

R2 v1 2026-06-23T01:34:43.280Z