Investigating Gender Stereotypes in Large Language Models via Social Determinants of Health
This addresses bias in LLMs for healthcare applications, but it is incremental as it builds on existing benchmarks by adding interaction analysis.
The study investigated gender bias in large language models by analyzing interactions between gender and social determinants of health in French patient records, finding that models rely on embedded stereotypes for gendered decisions.
Large Language Models (LLMs) excel in Natural Language Processing (NLP) tasks, but they often propagate biases embedded in their training data, which is potentially impactful in sensitive domains like healthcare. While existing benchmarks evaluate biases related to individual social determinants of health (SDoH) such as gender or ethnicity, they often overlook interactions between these factors and lack context-specific assessments. This study investigates bias in LLMs by probing the relationships between gender and other SDoH in French patient records. Through a series of experiments, we found that embedded stereotypes can be probed using SDoH input and that LLMs rely on embedded stereotypes to make gendered decisions, suggesting that evaluating interactions among SDoH factors could usefully complement existing approaches to assessing LLM performance and bias.