Student Hiring
I am currently hiring PhD students for Fall 2026 in the areas of Generative/Foundational AI and Medical Imaging.
Research Interests
ML algorithms
- Generative AI: diffusion models, vision-language models, synthetic data
- AI Safety: jailbreaking and misalignment of vision-lanugae models
- Private ML: differential privacy, bilevel optimization
- Robust ML: distribution shift, test-time adaptation, transferability
ML Applications
- Materials design: generative models, Bayesian optimization
- Health sciences: synthetic data generation, medical image analysis
Publications
You can find the list of publications from Google scholar.
News
Our paper titled "Seeing Clearly, Reasoning Confidently: Plug-and-Play Remedies for Vision Langugae Model Blindness" has been accepted for publication at CVPR (Feb 2026).
Our paper titled "Alternative Reprogramming for Service Models" has been accepted for publication at CVPR (Feb 2026).
Congrats to Yunbei for successfully passing the prospectus! (Feb 2026).
Our paper titled "Prompt-based Adaptation in Large-scale Vision Models: A Survey" has been accepted for publication at TMLR (Jan 2026).
A new graduate course CMPS 6820: Deep Learning will be offered soon. This course will cover topics from the basic models to modern tranformers and diffusion models.
Our paper titled "Doctor Approved: Generating Medically Accurate Skin Disease Images through AI-Expert Feedback" has been accepted for publication at NeurIPS'25 (Sep 2025).
Our paper titled "Visual Instance-aware Prompt Tuning" has been accepted for publication at ACM Multimedia'25 (Oct 2025).
Our paper titled "eSkinHealth: A Multimodal Dataset for Neglected Tropical Skin Diseases" has been accepted for publication at ACM Multimedia'25 (Oct 2025).
Our paper titled "SOFA: Deep Learning Framework for Simulating and Optimizing Atrial Fibrillation Ablation" has been accepted for publication at MICCAI'25 (Sep 2025).
Our paper titled "SoK: Can Synthetic Images Replace Real Data? A Survey of Utility and Privacy of Synthetic Image Generation" has been accepted for publication at USENIX'25 (Aug 2025).
Congrats to Tianyi for successfully passing the thesis defense! (July 2025)
Congrats to Ziyu for successfully passing the oral qualifier! (July 2025)
Our paper titled "DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation" has been accepted for publication at ICML'25 (July 2025).
Yilun Zhou is joining the machine learning group. Welcome!
Congrats to Xin for successfully passing the oral qualifier! (June 2025)
Congrats to Hajar for successfully passing the prospectus! (May 2025)
Congrats to Jiarui for successfully passing the oral qualifier! (Apr 2025)
Congrats to Henger for successfully defending the phd theses! (Feb 2025)
