Zhaohan Xi
I am an Assistant Professor of School of Computing at the Binghamton University, The State University of New York (SUNY), where I jointly contribute to the Computer Science and Information Systems programs. I am also an Affiliate Member of Upstate Cancer Center, Upstate Medical University, where I collaborate on research in cancer prevention, survivorship, therapeutics, and population health.
Before joining in Binghamton, I earned my Ph.D. at Pennsylvania State University, College of Information Science and Technology. I received my master’s degree from Lehigh University and my bachelor’s degree from Nanjing University of Aeronautics and Astronautics.
My research focuses on developing cross-disciplinary solutions to cutting-edge problems through agentic AI, aiming to foster responsible, robust, and resilient systems. On the methodology side, my research advances the reasoning, planning, and collaborative capabilities of LLM-based agents. I also work on techniques that improve their safety, trustworthiness, and reliability. On the application side, I bring these methods to domains where correctness and accountability matter most. My work spans healthcare, cybersecurity, AI-powered education, and scientific discovery, where I study how agentic systems can support expert decision making, strengthen digital defenses, personalize learning, and accelerate the pace of research.
Contact
School of Computing, Thomas J. Watson College of Engineering and Applied Science
Binghamton University, State University of New York
Engineering Building, EB G06
495 West Drive, Vestal, NY 13850
Email: zxi1n@binghamton.edu
news
| Jul 24, 2026 | I’ve been awarded a $2,000 CCE grant to build an AI-assisted app for head and neck cancer patients, with UHS and Upstate Medical. |
|---|---|
| Apr 30, 2026 | Three papers accepted by ICML’2026! Subjects: (1) Multi-Agent in Digital Health, (2) Threat Hunting Benchmark, (3) Spotlight (2.2% acceptance rate)Multi-Agent Uncertainty & Reinforcement Learning. |
| Jan 26, 2026 | One paper accepted by ICLR’2026! Subject: LLM Counterfactual Reasoning. |
selected publications
- TheraMind: a multi-LLM ensemble for accelerating drug repurposing in lung cancer via case report miningnpj Precision Oncology, 2026
- The value of variance: Mitigating debate collapse in multi-agent systems via uncertainty-driven policy optimizationIn Proceedings of the 43rd International Conference on Machine Learning, 2026
- Small agent group is the future of digital healthIn Proceedings of the 43rd International Conference on Machine Learning, 2026
- On the eligibility of LLMs for counterfactual reasoning: a decompositional studyIn The Fourteenth International Conference on Learning Representations, 2026