Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
Published over 50 papers in top journals and conferences such as IEEE TPAMI and CVPR, with more than 5,400 citations on Google Scholar. Research results have been reported by Forbes and other media, and he has been invited to give talks by top research teams like DeepMind. He has won several international academic competition championships, including the ImageNet competition, and the PREMIA Best Student Paper Award. The models he developed have been officially integrated into the Keras framework. He has led the development of over 20 achievements that have been commercially applied in more than ten flagship products/services of ByteDance, serving tens of millions of users daily. He has filed nearly 60 patents in China and the US, with 26 already granted.
Research Experience
Former founding member and technical leader of ByteDance’s US Research Institute, and has worked at Google Research and Snap Research in the United States.
Education
Graduated from the National University of Singapore in 2018; Master's and Bachelor's degrees were obtained from the Institute of Automation, Chinese Academy of Sciences (recommended) and Xidian University, respectively.
Background
Currently a professor and doctoral supervisor at the School of Computer Science, Beijing Jiaotong University. A recipient of the national high-level young talent program. Research interests include multimodal intelligence (perception, reasoning, and decision-making), world models, generative visual models, high-performance and efficient deep models, as well as AGI/ASI.
Miscellany
Organizer of international academic workshops such as FG2024 HAU and CVPR 2023 CVinW, and area chair for top conferences like NeurIPS. Recognized as a mentor for the Beijing Jiaotong University-Tencent Joint PhD Program, welcoming applications from recommended 2026 PhD students. Continuously recruiting postdocs, visiting students, and undergraduate interns.