mingyulu[at]cs[dot]washington[dot]edu.
About me
Hi! I am a final-year Computer Science & Engineering (CSE) Ph.D. student at the University of Washington, advised by Su-In Lee in the AIMS group. My research focuses on developing reliable, interpretable, and effective AI systems for biomedicine. Previously, I collaborated with Li-wei Lehman, Zach Shahn, and Finale Doshi-Velez, and interned at the MIT Laboratory for Computational Physiology, Bosch Research, and Microsoft AI (formerly MSR-RWE). Before my Ph.D., I earned an MD from Kaohsiung Medical University and a Master’s in Biomedical Informatics from Harvard Medical School.
My current research spans three directions:
- Data contributor attribution for diffusion models: I develop methods to trace model behavior back to influential training data, enabling better understanding, debugging, and auditing of diffusion models.
- Interpreting biological and treatment effects with language: I develop methods that use language to represent and explain complex biological and treatment effects.
- CellCLIP [NeurIPS 2025]
- ALEX [medRxiv]
- Improving and benchmarking LLM agents for biomedicine: I develop and evaluate LLM-based agents for biomedical tasks, with a focus on reliability, reasoning, and realistic healthcare workflows. I also contribute to the MMBU Challenge on benchmarking multimodal biomedical understanding.
Across these directions, I am particularly interested in translating advances in AI into real-world impact in healthcare and drug discovery. I have been fortunate to work with many wonderful collaborators and am always excited to connect about new research opportunities.
Recent update
- (June 2026) I will be working with Peniel Argaw and Tristan Naumann at Microsoft Research during the summer of 2026!
- (May 2026) New paper Agents that Matter: Optimizing Multi-Agent LLMs via Removal-Based Attribution!
- (April 2026) SurrogateSHAP has been accepted to ICML 2026!
- (January 2026) New paper SurrogateSHAP: Training-Free Contributor Attribution for Text-to-Image (T2I) Models!
- (September 2025) CellCLIP – Learning Perturbation Effects in Cell Painting via Text-Guided Contrastive Learning has been accepted to NeurIPS 2025!
- (June 2025) I will be working with Suraj Srinivas and Jorge Piazentin Ono at Bosch Research during the summer of 2025!
Conference publications
-
SurrogateSHAP: Training-Free Contributor Attribution for Text-to-Image (T2I) Models
Mingyu Lu, Soham Gadgil, Chris Lin, Chanwoo Kim, Su-In Lee
International Conference on Machine Learning (ICML), 2026
[paper] [code] -
CellCLIP – Learning Perturbation Effects in Cell Painting via Text-Guided Contrastive Learning
Mingyu Lu, Ethan Weinberger, Chanwoo Kim, Su-In Lee
Neural Information Processing Systems (NeurIPS), 2025
[paper] [code] [project page] -
BehaviorSFT: Behavioral Token Conditioning for Clinical Agents Across the Proactivity Spectrum
Yubin Kim, Zhiyuan Hu, Hyewon Jeong, Eugene W Park, Shuyue Stella Li, Chanwoo Park, Shiyun Xiong, MingYu Lu, Hyeonhoon Lee, Xin Liu, Daniel McDuff, Cynthia Breazeal, Samir Tulebaev, Hae Won Park
Findings of Empirical Methods in Natural Language Processing (EMNLP), 2025
[paper] [code] [project page] -
An Efficient Framework for Crediting Data Contributors of Diffusion Models
Chris Lin*, Mingyu Lu*, Su-In Lee
International Conference on Learning Representations (ICLR), 2025
[paper] [code] [project page] -
Learning to Maximize Mutual Information for Dynamic Feature Selection
Ian Connick Covert, Wei Qiu, MingYu Lu, Na Yoon Kim, Nathan J White, Su-In Lee
International Conference on Machine Learning (ICML), 2023
[paper] [code] -
G-Net: A Deep Learning Approach to G-computation for Counterfactual Outcome Prediction Under Dynamic Treatment Regimes.
Rui Li, Stephanie Hu, Yuria Utsumi, MingYu Lu, Prithwish Chakraborty, Daby Sow, Piyush Madan, Jun Li, Mohamed Ghalwash, Zachary Shahn, Li-wei H Lehman
Machine Learning for Health (ML4H), 2021
[paper] -
Is Deep Reinforcement Learning Ready for Practical Applications in Healthcare? A Sensitivity Analysis of Duel-DDQN for Sepsis Treatment.
MingYu Lu, Zach Shah, Finale Doshi Velez, Li-Wei Lehman.
American Medical Informatics Association (AMIA), 2020
Distinguished Paper
[paper]
Workshop and preprints
-
HealthAgentBench: A Unified Benchmark Suite of Realistic Agentic Healthcare Environments for Challenging Frontier AI Agents Qianchu Liu, Sheng Zhang, Guanghui Qin, Jeya Maria Jose Valanarasu, Maximilian Rokuss, Mingyu Lu, Timothy Ossowski, Juan Manuel Zambrano Chaves, Cliff Wong, Peniel Argaw, Yashna Hasija, Mu Wei, Wen-wai Yim, Qin Liu, Zilin Jing, Jason Entenmann, Naoto Usuyama, Tristan Naumann, Hoifung Poon
arXiv, 2026
[paper] [project page] -
Agents that Matter: Optimizing Multi-Agent LLMs via Removal-Based Attribution
Mingyu Lu*, YuShan Huang*, Su-In Lee
Agents in the Wild: Safety, Security, and Beyond at ICLR, 2026
[paper] -
ALEX: Automatic Language EXplanations for Interpreting Treatment Effects via Multi-Agents
Mingyu Lu, Chanwoo Kim, Nathan J. While, Su-In Lee
medRxiv, 2026
[paper] -
MedBLINK:Probing Visual Perception and Trustworthiness in Multimodal Language Models for Medicine
Mahtab Bigverdi, Wisdom Oluchi Ikezogwo, Kevin Minghan Zhang, Hyewon Jeong, MingYu Lu, Sungjae Cho, Linda Shapiro, Ranjay Krishna
Computer Vision for Automated Medical Diagnosis (CVAMD) at ICCV, 2025
Oral presentation
[paper] [project page] [hf dataset] -
Tiered Agentic Oversight: A Hierarchical Multi-Agent System for AI Safety in Healthcare
Yubin Kim, Hyewon Jeong, Chanwoo Park, MingYu Lu, Eugene W Park, Haipeng Zhang, Xin Liu, Hyeonhoon Lee, Daniel McDuff, Cynthia Breazeal, Samir Tulebaev, Hae Won Park
Multi-Agent Systems (MAS) in the Era of Foundation Models at ICML, 2025
[paper] -
Medical Hallucination in Foundation Models and Their Impact on Healthcare
Yubin Kim, Hyewon Jeong, Shan Chen, Shuyue Stella Li, Mingyu Lu, Kumail Alhamoud, Jimin Mun, Cristina Grau, Minseok Jung, Rodrigo Gameiro, Lizhou Fan, Eugene Park, Tristan Lin, Joonsik Yoon, Wonjin Yoon, Maarten Sap, Yulia Tsvetkov, Paul Liang, Xuhai Xu, Xin Liu, Daniel McDuff, Hyeonhoon Lee, Hae Won Park, Samir Tulebaev, Cynthia Breazea
medRxiv, 2025
[paper] [code] [project page] -
LIFT - XAI: Leveraging Important Features in Treatment Effects to Inform Clinical Decision-Making via Explainable AI
Mingyu Lu, Chanwoo Kim, Ian Covert, Nathan J. White, Su-In Lee
medRxiv, 2024
[paper] -
A Deep Bayesian Bandits Approach for Anticancer Drug Screening: Exploration via Functional Prior.
MingYu Lu, Yifang Chen, Su-In Lee
Adaptive Experimental Design and Active Learning in the Real World Workshop at ICML, 2022
[paper]
Leadership/Awards
| Year | Leadership/Awards |
|---|---|
| 2020 | Organizer of NewInML at NeurIPS 2020 |
| 2019 | LEAP Fellowship of the Ministry of Science and Technology of Taiwan. |
| 2017 | CoFounder of TinyNote |
Academic service
- Conference reviewing: NeurIPS (2024-2026), ICLR (2026), ICML (2026), ECCV (2026), AMIA (2020), CHIL (2020), ML4H@NeurIPS (2019,2020), DPFM@ICLR (2025,2026), LMRL@ICLR (2025)