I am currently a Professor in the College of Cyberspace Science and Technology, Beijing University of Technology (BJUT). Before joining BJUT, I was an assistant professor at the State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences (CASIA). I received my Ph.D. from USTC and CASIA in June 2024. My PhD supervisor is Prof. Changsheng Xu (IEEE/IAPR/CCF/CSIG Fellow). Before that, I obtained a Bachelor in telecommunication engineering from BUPT in June 2019.
(# denotes corresponding author, * denotes equal contribution)
Few-Shot Class-Incremental Learning with Low-Rank Prototype Adaptation
Jifei Luo*, Chaofan Chen*, Hantao Yao, Xiao Cui, Lu Yu, Changsheng Xu
Accepted by IEEE Transactions on Multimedia (TMM) 2026 (2026.08.12).
LAST: Bridging Vision-Language and Action Manifolds via Gromov-Wasserstein Alignment
Huaihai Lyu, Chaofan Chen#, Yuheng Ji, Xiansheng Chen, Pengwei Wang, Shanghang Zhang#, Changsheng Xu
Accepted by International Conference on Machine Learning (ICML) 2026.
General Covariant Action Modeling: Constructing Generalized Manifolds via Spatio-Temporal Decoupling
Huaihai Lyu, Chaofan Chen#, Mingyu Cao, Yuheng Ji, Changsheng Xu
Accepted by International Conference on Machine Learning (ICML) 2026.
SPDQ: Synergetic Prompts as Disentanglement Queries for Compositional Zero-Shot Learning
Han Jiang, Xiaoshan Yang, Chaofan Chen, Changsheng Xu
Accepted by IEEE Transactions on Multimedia (TMM) 2025.
EgoPrompt: Prompt Learning for Egocentric Action Recognition
Huaihai Lyu, Chaofan Chen#, Yuheng Ji, Changsheng Xu
Accepted by ACM Multimedia (ACM MM) 2025.
DMC³: Dual-Modal Counterfactual Contrastive Construction for Egocentric Video Question Answering
Jiayi Zou, Chaofan Chen#, Binkun Bao#, Changsheng Xu
Accepted by ACM Multimedia (ACM MM) 2025.