Examining Gender Bias in Languages with Grammatical Gender
Preprint
Abstract
Recent studies have shown that word embeddings exhibit gender bias inherited from the training corpora. However, most studies to date have focused on quantifying and mitigating such bias only in English. These analyses cannot be directly extended to languages that exhibit morphological agreement on gender, such as Spanish and French. In this paper, we propose new metrics for evaluating gender bias in word embeddings of these languages and further demonstrate evidence of gender bias in bilingual embeddings which align these languages with English. Finally, we extend an existing approach to mitigate gender bias in word embeddings under both monolingual and bilingual settings. Experiments on modified Word Embedding Association Test, word similarity, word translation, and word pair translation tasks show that the proposed approaches effectively reduce the gender bias while preserving the utility of the embeddings.Keywords
All Related Versions
This publication has 4 references indexed in Scilit:
- Evaluating the Underlying Gender Bias in Contextualized Word Embeddings2019
- Enriching Word Vectors with Subword InformationTransactions of the Association for Computational Linguistics, 2017
- Semantics derived automatically from language corpora contain human-like biasesScience, 2017
- THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMSAnnals of Eugenics, 1936