PhD student at Georgetown University (GUCV Lab), advised by Prof. Sarah Adel Bargal.
I build diagnostic frameworks that pinpoint where and why vision-language models fail on fine-grained image and video understanding. My work spans multimodal hallucination, temporal reasoning in video, and probe-based evaluation. The goal: make visual reasoning traceable and verifiable, instead of judging a model only by its final answer.
🔍 Looking for a Summer 2027 research internship in multimodal and video understanding. If you work on this, I'd love to talk: mk2524@georgetown.edu