EuroGEST: Investigating gender stereotypes in multilingual language models

TL;DR

EuroGEST is a multilingual benchmark dataset that measures gender stereotypes in large language models across 30 European languages, revealing persistent stereotypes and the influence of model size and fine-tuning.

Contribution

This work introduces EuroGEST, a novel multilingual dataset for assessing gender bias in LLMs, expanding existing benchmarks with translation and heuristics, and providing insights into stereotype encoding across languages.

Findings

Larger models encode stereotypes more strongly.

Instruction fine-tuning does not reliably reduce stereotypes.

Women are stereotypically associated with 'beautiful', 'empathetic', 'neat'.

Abstract

Large language models increasingly support multiple languages, yet most benchmarks for gender bias remain English-centric. We introduce EuroGEST, a dataset designed to measure gender-stereotypical reasoning in LLMs across English and 29 European languages. EuroGEST builds on an existing expert-informed benchmark covering 16 gender stereotypes, expanded in this work using translation tools, quality estimation metrics, and morphological heuristics. Human evaluations confirm that our data generation method results in high accuracy of both translations and gender labels across languages. We use EuroGEST to evaluate 24 multilingual language models from six model families, demonstrating that the strongest stereotypes in all models across all languages are that women are 'beautiful', 'empathetic' and 'neat' and men are 'leaders', 'strong, tough' and 'professional'. We also show that larger models encode gendered stereotypes more strongly and that instruction finetuning does not consistently reduce gendered stereotypes. Our work highlights the need for more multilingual studies of fairness in LLMs and offers scalable methods and resources to audit gender bias across languages.