Large Language Models for Medicine
Pioneering the application of large language models to medical domains, from clinical documentation and literature mining to patient-trial matching and medical coding.
This research area explores how large language models can be adapted and specialized for medical applications, addressing unique challenges in healthcare such as domain-specific terminology, regulatory compliance, and clinical workflow integration.
The work encompasses developing medical LLMs that can understand clinical contexts, generate accurate clinical documentation, and assist in complex medical decision-making processes. Key innovations include foundation models for medical literature analysis, specialized architectures for clinical trial matching, diagnosis-related group prediction systems, and process-supervised reward models for clinical note generation.
The research tackles critical challenges including medical knowledge grounding, hallucination prevention in high-stakes medical contexts, and the development of human-AI collaborative systems that enhance rather than replace clinical expertise.
A Foundation Model for Human-AI Collaboration in Medical Literature Mining (arXiv 2025)
Matching Patients to Clinical Trials with Large Language Models (Nature Communications, 2024)
DRG-LLaMA: Tuning LLaMA Model to Predict Diagnosis-Related Group for Hospitalized Patients (npj Digital Medicine, 2024)
Process-Supervised Reward Models for Clinical Note Generation (arXiv 2024)
Interested in This Research Area?
We welcome collaborations with researchers, clinicians, and industry partners working in large language models for medicine. Our lab is always looking for motivated students and postdocs to join our team.