Hansheng Zeng
Hansheng Zeng

Hansheng Zeng 曾瀚圣

Junior Computer Science Student · The University of Hong Kong

Hi, I am Hansheng Zeng, a junior studying Computer Science at The University of Hong Kong. I was a visiting student in Computer Science through Berkeley Global Access at University of California, Berkeley in Spring 2026, and attended the GLOBEX Julmester Program at the College of Engineering, Peking University in Summer 2024. As a Visiting Undergraduate Scholar in the Department of Computer Science at Johns Hopkins University, I work on multimodal learning with Prof. Alan Yuille at the Computational Cognition, Vision, and Learning Group. At The University of Hong Kong, I work on AI agents with Prof. Chao Huang at the Data Intelligence Lab@HKU. I also work on spatiotemporal prediction with Prof. Chuanguang Yang at the Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS). My research interests include multimodal learning, AI agents, and spatiotemporal prediction.

Email Google Scholar GitHub LinkedIn DBLP

News

  • My first-author paper was submitted to AAAI 2027.

Selected Publications

Other publications ↓

* Equal contribution.

Architecture of MAD, integrating Mamba modules and Diffusion Transformers for spatiotemporal prediction
First Author

Enhancing Spatiotemporal Prediction through the Integration of Mamba State Space Models and Diffusion Transformers

Hansheng Zeng, Yuqi Li, Ruize Niu, Chuanguang Yang, Shiping Wen

Knowledge-Based Systems (KBS), 2025

[paper] [code]

We integrate Mamba state-space modeling with Diffusion Transformers for accurate and robust spatiotemporal prediction.

Architecture of SDKD for frequency-aligned knowledge distillation in lightweight spatiotemporal forecasting
ICCV 2025

Frequency-Aligned Knowledge Distillation for Lightweight Spatiotemporal Forecasting

Yuqi Li, Chuanguang Yang, Hansheng Zeng, Zeyu Dong, Zhulin An, Yongjun Xu, Yingli Tian, Hao Wu

IEEE/CVF International Conference on Computer Vision (ICCV), 2025

[paper] [code]

We transfer frequency-aligned, multi-scale knowledge from complex teachers to efficient forecasting models.

Overview of reinforcement learning-driven k-space sampling, MRI reconstruction, and medical image segmentation
Co-first Author

Efficient Medical Image Segmentation via Reinforcement Learning-Driven K-Space Sampling

Yuqi Li*, Hansheng Zeng*, Fuyan Zhang, Chuanguang Yang, Yanli Li, Weiping Ding

IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI), 2026

[paper] [code]

We learn an adaptive k-space sampling policy that balances MRI acquisition efficiency and segmentation quality.

Other Publications

Google Scholar ↗
  1. DIEq: Dynamic Identity Equilibrium for Author Disambiguation in KDD Cup 2024 WhoIsWho-IND Challenge

    Zhixiang Lu, Hansheng Zeng, Yuqi Li

    KDD 2024 OAG-Challenge Cup Workshop, 2024 Team ranked 10th out of 365 teams

  2. Towards Robust Medical Image Segmentation: Spectro-Spatial Domain Generalization with MRAM and DMIR

    Yuqi Li, Junhao Dong, Hansheng Zeng, Fuyan Zhang, Zeyu Dong, Chuanguang Yang, Yingli Tian

    Computer Vision and Image Understanding, 2026

  3. Evolving Multimodal Models for Physical Dynamics: A Multi-Objective Neuroevolution Approach

    Yuqi Li, Junhao Dong, Jiao Liu, Piotr Koniusz, Hansheng Zeng, Chuanguang Yang, Junming Liu, Yingli Tian, Tingwen Huang, Hao Wu

    IEEE Transactions on Evolutionary Computation, 2026

  4. AMMKD: Adaptive Multimodal Multi-Teacher Distillation for Lightweight Vision-Language Models

    Yuqi Li, Chuanguang Yang, Junhao Dong, Zhengtao Yao, Haoyan Xu, Zeyu Dong, Hansheng Zeng, Zhulin An, Yingli Tian

    arXiv preprint, 2025

  5. CPG: Channel Pruning with DFS Guided Grouping for Efficient Medical Image Segmentation

    Xilin Yan, Fuyan Zhang, Yuqi Li, Zhen Qiu, Boyuan Zhao, Jiarui Li, Hansheng Zeng, Yihao Chen, Chuanguang Yang

    International Conference on Neural Information Processing (ICONIP), 2024