Run Like a Girl! Sports-Related Gender Bias in Language and Vision

TL;DR

This paper investigates gender bias in language and vision datasets and models, revealing underrepresentation of women and biased naming practices that reinforce stereotypes, with implications for harmful societal impacts.

Contribution

It provides a comprehensive analysis of gender bias in datasets and models, highlighting underrepresentation and biased naming, and demonstrates how these biases perpetuate stereotypes and harm.

Findings

Women are underrepresented in datasets, promoting invisibilization.

Speakers produce more sport-related names for men than women (46% vs. 35%).

A model trained on naming data reproduces gender bias.

Abstract

Gender bias in Language and Vision datasets and models has the potential to perpetuate harmful stereotypes and discrimination. We analyze gender bias in two Language and Vision datasets. Consistent with prior work, we find that both datasets underrepresent women, which promotes their invisibilization. Moreover, we hypothesize and find that a bias affects human naming choices for people playing sports: speakers produce names indicating the sport (e.g. 'tennis player' or 'surfer') more often when it is a man or a boy participating in the sport than when it is a woman or a girl, with an average of 46% vs. 35% of sports-related names for each gender. A computational model trained on these naming data reproduces the bias. We argue that both the data and the model result in representational harm against women.