← back to publications
Gender Bias in Contextualized Word Embeddings
publication

Gender Bias in Contextualized Word Embeddings

Jieyu Zhao, Tianlu Wang, Mark Yatskar, Ryan Cotterell, Vicente Ordonez, Kai-Wei Chang.
North American Chapter of the Association for Computational Linguistics. NAACL 2019. short. Minneapolis, Minnesota. June 2019.
paper arxiv pdf raw bibtex

abstract

In this paper, we quantify, analyze and mitigate gender bias exhibited in ELMo's contextualized word vectors. First, we conduct several intrinsic analyses and find that (1) training data for ELMo contains significantly more male than female entities, (2) the trained ELMo embeddings systematically encode gender information and (3) ELMo unequally encodes gender information about male and female entities. Then, we show that a state-of-the-art coreference system that depends on ELMo inherits its bias and demonstrates significant bias on the WinoBias probing corpus. Finally, we explore two methods to mitigate such gender bias and show that the bias demonstrated on WinoBias can be eliminated.

citation

@inproceedings{zhao2019gender,
  title = {Gender Bias in Contextualized Word Embeddings},
  author = {Zhao, Jieyu and Wang, Tianlu and Yatskar, Mark and Cotterell, Ryan and Ordonez, Vicente and Chang, Kai-Wei},
  year = {2019},
  booktitle = {North American Chapter of the Association for Computational Linguistics. NAACL 2019},
  url = {https://arxiv.org/abs/1904.03310},
}