I am a Ph.D. student in Computer Science at the University of Massachusetts Amherst. My research focuses on large language models, agentic retrieval, reasoning, and AI for biomedicine.
My primary research centers on agentic retrieval, reinforcement learning, and large language model pretraining for specialized domains, including high-stakes medical settings such as cancer, opioid overdose, and diagnosis prediction, as well as information retrieval, retrieval-augmented generation, and graph-based RAG frameworks. More broadly, I am interested in building agentic AI systems that can search, plan, retrieve evidence, and solve complex tasks reliably in high-stakes environments. A central goal of my work is to develop systems that are not only effective, but also more robust, interpretable, and reliable in open-ended settings. I am also interested in extending these methods to the medical domain, where accurate evidence gathering and reasoning are especially important.
My recent work spans several interconnected directions:
- Developing and studying large language model systems for multi-step reasoning, complex problem solving, and embodied AI.
- Improving agentic retrieval pipelines so models can retrieve the right evidence at the right time.
- Building tool-using LLM agents that are more robust, interpretable, and effective in open-ended environments.
Recent Preprints
Coming soon！Selected Publications
For the latest and complete publication list, please also visit my Google Scholar profile.
-
Efficient and Effective Internal Memory Retrieval for LLM-Based Healthcare Prediction.
ACL 2026 Finding (reviewer score 4,3.5, 3,5 but findings)
-
Cancerllm: A large language model in cancer domain
NPJ Digital Medicine
-
Science Advance
-
Biomedrag: A retrieval augmented large language model for biomedicine
Journal of Biomedical Informatics (JBI) 162, 104769
-
Journal of the American Medical Informatics Association (JAMIA) 32 (3), 545-554
-
Journal of Medical Internet Research 27, e70080
-
RT: a Retrieving and Chain-of-Thought framework for few-shot medical named entity recognition
Journal of the American Medical Informatics Association (JAMIA) 31 (9), 1929 ...
-
LEAP: LLM instruction-example adaptive prompting framework for biomedical relation extraction
Journal of the American Medical Informatics Association (JAMIA) 31 (9), 2010 ...
-
Journal of Biomedical Informatics (JBI)
-
A condensed transition graph framework for zero-shot link prediction with large language models
ICDM 2024
-
SIGIR 2023
-
A Review of Reinforcement Learning for Natural Language Processing, and Applications in Healthcare
Journal of the American Medical Informatics Association (JAMIA)
-
Semantic Structure based Query Graph Prediction for Question Answering over Knowledge Graph
COLING 2022
-
A Hierarchical N-Gram Framework for Zero-Shot Link Prediction
EMNLP 2022 (Finding track)
Experience
- PhD Student, Computer Science, UMass Amherst. 09.2024 - Present
- Researcher, University of Minnesota Twin Cities. 01.2023 - 08.2024
- Teaching Assistant, Georgia State University, Deep Learning (CSC 8850). 09.2021 - 12.2021
- Research Assistant, Georgia State University. 01.2021 - 05.2021
Service
- Program Committee: ACL 2023, IEEE ICHI 2023, EMNLP Industry Track 2023
- Reviewer: EMNLP 2022, EMNLP 2023, NAACL 2024, Artificial Intelligence in Medicine, Engineering Applications of Artificial Intelligence, Expert Systems with Applications, Information Processing & Management, International Journal of Medical Informatics, Npj Health Systems, Journal of Artificial Intelligence Research
- Other Service: Central Functions Committee in SCD 2022, Student Volunteer at EMNLP 2022
Contact
Google Scholar: scholar profile
GitHub: ToneLi
Email: mingchenli@umass.edu