EuroGEST: Investigating gender stereotypes in multilingual language models
This addresses fairness issues in AI for multilingual users, though it is incremental as it builds on existing benchmarks.
The authors tackled the problem of gender bias in multilingual language models by introducing EuroGEST, a dataset to measure gender-stereotypical reasoning across 30 languages, and found that models consistently encode stereotypes such as women being 'beautiful' and men being 'leaders', with larger models showing stronger biases.
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.