Yu Kong

Department of Computer Science and Engineering, Michigan State University

yukong.jpg

Office: EB 3573

Email: yukong@msu.edu

Dr. Yu Kong is an Assistant Professor of the Department of Computer Science and Engineering, Michigan State University. He is now directing the ACTION Lab. He serves Pattern Recognition and IEEE Transactions on Multimedia as an Associate Editor. He also serves CVPR, ICCV, IJCAI, AAAI, Multimedia, FG, etc., as an Area Chair. He is an IEEE Senior Member. His research has been supported by National Science Foundation, Office of Naval Research, Army Research Office, Google, Meta, Nvidia, Adobe, Josiah Mary Jr Foundation, and Knight Foundation.

Dr. Kong’s research focuses on reliable video intelligence and multimodal learning, with a particular emphasis on physical intelligence from human-centric videos. His research addresses fundamental challenges in learning physically grounded representations of humans and their interactions with objects and environments, modeling and generating physical interaction dynamics, integrating symbolic knowledge for reliable video understanding and reasoning, and transferring knowledge from human experiences to embodied intelligent systems.

Current research topics are

  • Human motion and human-object interaction modeling
  • Generative video modeling
  • Open world, multi-modal, self-supervised learning
  • Robot learning
  • Video understanding, ego-centric video understanding

Openings: We have openings for Fall 2027 for Ph.D. students, Visiting Scholars, and Research Assistants. You can either 1) fill in the external contact form (preferred) or 2) send your application package to actionlab@msu.edu. I will review applications periodically. Due to large volume of applications, I am not able to track and respond to every applicant. How to apply?

news

Oct 15, 2026 We have one papers accepted by WACV 2027! Congratulations to Yifan and collaborators!
Sep 24, 2026 We have six papers accepted by NeurIPS 2026! Congratulations to all my students and collaborators!
Jun 18, 2026 Our paper on test-time optimization for video generation has been accepted by ECCV 2026. Congratulations to Yujiang!
Apr 30, 2026 We have one paper on continual VQA accepted by ICML 2026.
Apr 02, 2026 One paper about media forenisics has been accepted by FG 2026. This is the Zhongyi’s first paper at ACTION Lab. Congratulations to Zhongyi and collaborators!
Feb 21, 2026 We have two papers accepted by CVPR 2026. Congratulations to Zhanbo and our collaborators!
Jan 26, 2026 One paper about mistake detection has been accepted by ICLR 2026. This is the Wenliang’s first paper at ACTION Lab. Congratulations to Wenliang and Yujiang!
Nov 10, 2025 One paper about Egocentric Video Generation has been accepted by WACV 2026. Congratulations to Yujiang!
Sep 25, 2025 Our project on Neuro-Symbolic Video Understanding ($50K) is selected for funding by Jenison Fund.
Sep 18, 2025 We have one paper accepted by NeurIPS 2025. Congratulations to Yifan!

selected publications

  1. NeurIPS
    Zhanbo2026c.pdf
    H-Flow: Self-supervised Human Scene Flow via Physics-inspired Joint Multi-modal Learning
    Zhanbo Huang, Xiaoming Liu, and Yu Kong
    In Neural Information Processing Systems (NeurIPS), 2026
  2. WACV
    Yujiang2025.pdf
    Show Me: Generating Instructional Videos with Diffusion Models
    Yujiang Pu, Zhanbo Huang, Vishnu Boddeti, and Yu Kong
    In Winter Conference on Applications of Computer Vision (WACV), 2026
  3. NeurIPS
    YifanNeurIPS2025.pdf
    IndustryEQA: Pushing the Frontiers of Embodied Question Answering in Industrial Scenarios
    Yifan Li, Yuhang Chen, Anh Dao, Lichi Li, Zhongyi Cai, Zhen Tan, Tianlong Chen, and 1 more author
    In NeurIPS DB Track, 2025
  4. CVPR Highlight
    Zhanbo2025.pdf
    Learning Human-centric Motion Representation for Action Analysis
    Zhanbo Huang, Xiaoming Liu, and Yu Kong
    In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025
    Highlight

acknowledgement of support