A Toolbox for Surfacing Health Equity Harms and Biases in Large Language Models

arXiv:2403.12025v1 Announce Type: cross
Abstract: Large language models (LLMs) hold immense promise to serve complex health information needs but also have the potential to introduce harm and exacerbate health disparities. Reliably evaluating equity-related model failures is a critical step toward developing systems that promote health equity. In this work, we present resources and methodologies for surfacing biases with potential to precipitate equity-related harms in long-form, LLM-generated answers to medical questions and then conduct an empirical case study with Med-PaLM 2, …

abstract arxiv biases cs.cl cs.cy cs.lg equity harm health health equity information language language models large language large language models llms promote serve systems type