Published papers at ICLR, NeurIPS, etc.; recognized as an Outstanding Reviewer for CVPR 2025; passed the PhD candidacy exam titled “Visual Test-Time Scaling via Search”.
Research Experience
Currently a PhD student at VILab, EPFL, previously worked on semantic segmentation, test-time adaptation, domain generalization, and uncertainty estimation at ShanghaiTech.
Education
PhD: École polytechnique fédérale de Lausanne (EPFL), supervised by Prof. Amir Zamir; Master's: ShanghaiTech, supervised by Prof. Xuming He.
Background
Research interests: Machine learning and computer vision, particularly exploring System-2–like thinking abilities in vision and multimodal models. Also curious about how System-1–style generalizable representation learning and System-2–style reasoning and adaptation can be effectively integrated.
Miscellany
Served as a reviewer for multiple international conferences such as CVPR, NeurIPS, and ICLR, and as a teaching assistant at both ShanghaiTech and EPFL.