This repository contains supplementary materials and data for the paper "PALLM: Evaluating and Enhancing Palliative Care Conversations with Large Language Models," which was submitted to the ACM Transactions on Computing for Healthcare’s Special Issue on Large Language Models, Conversational Systems, and Generative AI in Health.
The repository currently includes benchmark scripts (in scripts/) developed and annotated by our clinical team to assess language models' effectiveness in evaluating key clinical communication metrics: understanding, empathy, emotion, presence, and clarity.