🌴 About Me
I am a master’s student at Tsinghua University, under the supervision of Prof. Xiu Li. I’m now a visiting student at the LONG Group, Hong Kong University of Science and Technology (HKUST), working with Prof. Long Chen. I obtained my B.Eng in Computer Science and Technology at Harbin Institute of Technology (Shenzhen) in 2024.
My research interests lie in Computer Vision, particularly in visual content generation such as Concept Customization (MultiBooth), Customized Editing (InstantSwap).
I am actively seeking for research internship opportunities in either academia or industry. Please feel free to contact me via email.
🔥 News
-
Mar 2026Excited to release MoKus, A novel framework for knowledge-aware concept customization through cross-modal knowledge transfer, enabling robust and high-fidelity customized generation.
-
Sep 2025Join LONG Group@HKUST as a visiting student.
-
Jan 2025InstantSwap is accepted by ICLR2025.
-
Dec 2024MultiBooth is accepted by AAAI2025.
-
Dec 2024Excited to release InstantSwap, an efficient customized concept swapping method across sharp shape differences.
-
Apr 2024Excited to release MultiBooth, a novel and efficient multi-concept customization method.
-
Dec 2023Started to cooperate with Yue Ma.
-
Aug 2023Started to cooperate with Dr. Kai Li.
-
Jul 2023Awarded as an outstanding camper in Tsinghua University Summer Camp.
-
Apr 2023Started internship in Professor Li Xiu’s research group.
📝 Publications
Chenyang Zhu, Hongxiang Li, Xiu Li, Long Chen $^{\dagger}$
arXiv preprint arXiv:2603.12743, 2026
A novel framework for knowledge-aware concept customization through cross-modal knowledge transfer, enabling robust and high-fidelity customized generation.
Chenyang Zhu $^{*}$, Kai Li $^{*,\dagger}$, Yue Ma $^{*}$, Longxiang Tang, Chengyu Fang, Chubin Chen, Qifeng Chen, Xiu Li $^{\dagger}$
In International Conference on Learning Representations (ICLR), 2025
A novel training-free customized concept swapping framework, which enables efficient concept swapping across sharp shape differences.
Chenyang Zhu, Kai Li $^{\dagger}$, Yue Ma, Chunming He, Xiu Li $^{\dagger}$
In The AAAI Conference on Artificial Intelligence (AAAI), 2025
A novel and efficient technique for multi-concept customization in image generation from text.