Publications: Including EgoLife, Octopus, FunQA, Otter, LMMs-Eval, UPD, PSG, PVSG, PSG4D, OpenOOD, CoOp, CoCoOp, etc.; Awards: Outstanding Reviewer at CVPR-2025, CVPR-2024, and ECCV-2024; Conference Roles: Area Chair at ACL-25 and EMNLP-24, Outstanding Reviewer at CVPR-2025.
Research Experience
Co-founder of Synvo AI; During his Ph.D., he worked on large multimodal models (e.g., EgoLife, Octopus, FunQA, Otter), AI safety for foundation models (e.g., UPD), the PSG series (e.g., PSG, PVSG, PSG4D), OOD detection, and prompt tuning (e.g., CoOp, CoCoOp).
Education
Ph.D.: 2021-2025, Nanyang Technological University (NTU) CCDS, MMLab@NTU, supervised by Dr. Ziwei Liu; B.S.: 2017, joint program between Beijing University of Posts and Telecommunications (BUPT) and Queen Mary University of London (QMUL).
Background
Research Interests: Multimodal models, contextual intelligence, and AI safety. Professional Field: Computer Vision, Natural Language Processing. Brief Introduction: Co-founded Synvo AI, focusing on promoting the co-evolution of AI and humans through contextual intelligence and multimodal models.
Miscellany
Personal Interests: Attending academic conferences and sharing research results, such as carrying a shiny backpack at ECCV-22; Organizing events: The AI Talks; Social Media: Twitter, Facebook, GitHub.