Published multiple papers across top conferences such as CVPR, NeurIPS, ICLR, and ECCV; involved in or led several projects like CLIPS, Story-Adapter, MedTrinity-25M, etc.; members have served as area chairs for important international conferences including ICML 2025, CVPR 2025, and ICCV 2025.
Research Experience
Focused on developing efficient deep learning models that work with minimal supervision; exploring how to maintain model performance when facing (adversarial) data distribution shifts; committed to creating transparent and trustworthy medical AI systems that can function effectively in complex clinical environments; also conducting research into generative AI, including large language models (LLMs) and diffusion models.
Background
Research interests span computer vision and machine learning, with a focus on developing efficient deep representation learning with minimal supervision, securing model performance under (adversarial) distribution shifts, building transparent and trustworthy medical AI systems, and generative AI including LLMs and Diffusion Models.
Miscellany
Lab members have received various honors, such as the Jack Baskin & Peggy Downes-Baskin Fellowship; also achieved second place in the NeurIPS 2023 Trojan Detection Challenge.