Lab
Research Interests
AI for Health
Clinical Informatics
Personalized Medicine
Large Language Models
Agentic AI
Tool-Augmented Learning
Retrieval-Augmented Learning
Reasoning
Data-Centric AI
-
Tool-Augmented LLM Agents for Clinical Research and Biomedical Discovery
Developing LLM agents that perform structured question interpretation, retrieval-augmented planning, governed tool selection, code synthesis, sandboxed execution, and self-verification with clinician checkpoints. Building agentic training environments with reinforcement learning so agents practice safe tool use, recover from execution errors, and improve from human and environmental feedback.
-
Reasoning-Aware Retrieval-Augmented LLMs for Evidence-Based Medicine
Unifying reasoning and retrieval so that LLM outputs are auditable and decision-ready. Formalizing reasoning-aware retrieval where agents decompose clinical questions, plan evidence needs, retrieve and ground each step, and verify intermediate claims through self-play and collaborative supervision. Developing domain-tuned biomedical retrievers and EHR-aware augmentation that supply trustworthy context with citations, rationales, and uncertainty signals.
-
Responsible AI for Transparent and Trustworthy Clinical Models
Ensuring that AI systems for clinical decision support are trustworthy, explainable, and fair. Addressing challenges of data quality, interpretability, and bias in medical AI through fairness-aware modeling, knowledge-infused data generation, and interpretable attention mechanisms to establish accountable and equitable AI deployment in healthcare.
-
AI-Powered Precision Medicine for Personalized Patient Care
Advancing AI methods to support precision medicine by combining clinical data with causal inference techniques to tailor predictions and treatments to each patient. Developing models for individualized outcome prediction, personalized causal graph learning, and NLP techniques to extract social and behavioral determinants of health from clinical notes.
Lab Members
- Junhui Mi, Ph.D. Student in Biostatistics, UTSW
- Yi Jiang, Ph.D. Student in Biomedical Engineering, UTSW (co-advised w/ Dr. Yang Xie)
- Jinrui Fang, Ph.D. Student, UT Austin (co-advised w/ Drs. Ying Ding & Yang Xie)
Recent Research Talks
- Panel Discussion, Text Mining Community of Special Interest at ISMB 2026, Washington, D.C., July 2026
"Trustworthy literature agents for biomedical discovery: grounding, reproducibility, and evaluation"
- Tutorial at ISMB 2026, Washington, D.C., July 2026
"Large Language Models and Agentic AI for Biomedical Informatics"
- Tutorial at ACL 2026, San Diego, CA, July 2026
"Towards Effective and Efficient Multi-Agent Language Model Systems: Foundations, Prospects, and Applications"
- Invited Talk at NeurIPS 2025 CURE-Bench Data Competition, San Diego, CA, Dec 2025
- Invited Keynote Talk at Annual Biomedical STAR-AI Workshop, Georgia Tech & Emory University, Atlanta, GA, Oct 2025
- Invited Talk at UT Southwestern Medical School New Faculty Research Forum, Dallas, TX, Oct 2025
- Invited Talk at Stanford MedAI Group Exchange Sessions, Palo Alto, CA, Aug 2025
"MedAgentGym: Training LLM Agents for Code-Based Medical Reasoning at Scale"