About
I am now a fourth-year Ph.D. student in VMC group under the supervision of Professor Shiliang Zhang at the School of Computer Science of Peking University, Beijing, China.
My research interests are multi-modal understanding and generation, including multimodal large language model, image/video generation, and open-vocabulary recognition. I am seeking job opportunities in 2026. Please feel free to email me if you are interested in my research.
News
🎉 MagCache about fast video generation has been accepted by NeurIPS 2025
🎉 EMLoC about long-context learning has been accepted by ICML 2025
🎉 MMRef about multi-modal representation learning has been accepted by IEEE TMM
🎉 OVMR about open-vocabulary recognition has been accepted by CVPR 2024
Education
Peking University
2022 - PresentPh.D. in School of Computer Science, Supervised by Prof. Shiliang Zhang
Northwestern Polytechnical University
2018 - 2022BSc in School of Software. Practice Research Advised by Prof. Peng Wang
Publications
MagCache: Fast Video Generation with Magnitude-Aware Cache
Zehong Ma, Longhui Wei, Feng Wang, Shiliang Zhang, Qi Tian
Neural Information Processing Systems (NeurIPS) 2025
DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation
Zehong Ma, Longhui Wei, Shuai Wang, Shiliang Zhang, Qi Tian
2025
Multi-Modal Reference Learning for Fine-Grained Text-to-Image Retrieval
Zehong Ma, Hao Chen, Wei Zeng, Limin Su, Shiliang Zhang
IEEE Transactions on Multimedia 2025
Honors and Awards
Top Ten Students of the Year
2025NERCVT, Peking University
Merit Student
2025Peking University
China National Scholarship
2019,2020,2021Outstanding Student Model of Northwestern Polytechnical University
2020National Champion of China Robotics Competition in Basketball Robot
2020National First Prize of DJI RoboMaster Competition
2020Services
Reviewer
2023 - PresentNeurIPS, CVPR, TIP, TMM, TMLR, CVIU