Video Understanding · Agentic AI · World Models

Haonan Ge葛浩南

Ph.D. Student in Computer Science, UC Santa Barbara

johnny.jpg

I am a CS Ph.D. student at the University of California, Santa Barbara, advised by Prof. Yao Qin and Prof. Zhen Dong. I received my bachelor's degree in Electrical and Computer Engineering from Southeast University.

Before joining UCSB, I was a Research Intern with UC Merced and The University of Queensland, advised by Prof. Yujun Cai (UQ) and Prof. Yiwei Wang (UC Merced), and I continue to collaborate with them as well as Prof. Kai-Wei Chang (UCLA) and Prof. Ming-Hsuan Yang (UC Merced).

haonange@ucsb.edu · gehaonan82@gmail.com

I am currently seeking research internship opportunities for Summer 2027. I am also always open to research collaborations, and students who would like to intern with me are very welcome. Feel free to reach out by email anytime!

Research Interests

01

Video Understanding and Memory for Multimodal LLMs

I build multimodal LLMs that perceive, remember, and reason over long and streaming video: deciding what to keep in memory, which frames to look at, and how to ground answers in visual evidence rather than language priors.

02

Embodied Agents and World Models

I study agents that turn perception into action: robot-use agents that learn which way of acting (direct motion, motion planning, control programs, or VLA policies) they can execute reliably and cheaply, and world models that let them anticipate the consequences of their actions.

News

Nov 13, 2026 I will be attending NeurIPS 2026 in Atlanta, Georgia, from December 8 to 13. See you in Atlanta! :wave:
Sep 24, 2026 “What Should a Streaming Video Model Remember?” has been accepted to NeurIPS 2026 ! :tada:
Sep 24, 2026 I have officially joined University of California, Santa Barbara (UCSB) as a CS Ph.D. student! Feel free to reach out for collaborations or research internships.
Apr 05, 2026 I have accepted the Reviewer invitation for NeurIPS 2026.
Apr 01, 2026 I have accepted the Ph.D. offer from University of California, Santa Barbara (UCSB).
Jan 26, 2026 “SportR: A Benchmark for Multimodal Large Language Model Reasoning in Sports” has been accepted to ICLR 2026, See you in Rio de Janeiro!
Oct 02, 2025 I have been invited as a Reviewer in CVPR 2026 ! :sparkles: :smile:
Sep 25, 2025 Three Paper have been submitted to ICLR 2026 !
Aug 21, 2025 “MRFD: Multi-Region Fusion Decoding with Self-Consistency for Mitigating Hallucinations in LVLMs” has been accepted to EMNLP 2025 !
Jun 25, 2025 I start a Summer Research Intern in UCI and Rice University.

Selected Publications

  1. NeurIPS
    selectstream.png
    What Should a Streaming Video Model Remember?
    Haonan Ge, Yiwei Wang, Hang Wu, and 1 more author
    In NeurIPS, 2026
  2. EMNLP
    guigame.png
    GUI Agents for Continual Game Generation
    Yixu Huang, Bo Li, Na Li, and 8 more authors
    In EMNLP, 2026
  3. NeurIPS
    camreasoner.png
    CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
    Hang Wu, Yujun Cai, Zehao Li, and 4 more authors
    In NeurIPS, 2026
    Under review.
  4. NeurIPS
    Framemind.png
    FrameMind: Frame-Interleaved Video Reasoning via Reinforcement Learning
    Haonan Ge, Yiwei Wang, Kai-Wei Chang, and 2 more authors
    In NeurIPS, 2026
    Under review.
  5. ICLR
    SPORTR.png
    SportR: A Benchmark for Multimodal Large Language Model Reasoning in Sports
    Haotian Xia*, Haonan Ge*, Junbo Zou*, and 16 more authors
    In ICLR, 2026
  6. arXiv
    Refineshot.png
    RefineShot: Rethinking Cinematography Understanding with Foundational Skill Evaluation
    Hang Wu, Yujun Cai, Haonan Ge, and 3 more authors
    arXiv preprint arXiv:2510.02423, 2025
  7. EMNLP
    MRFD.png
    MRFD: Multi-Region Fusion Decoding with Self-Consistency for Mitigating Hallucinations in LVLMs
    Haonan Ge, Yiwei Wang, Ming-Hsuan Yang, and 1 more author
    In EMNLP, 2025

* denotes equal contribution

Academic Services

Conference Reviewer ICLR 2027·WACV 2027·NeurIPS 2026·ACM MM 2026·CVPR 2026·WACV 2026 Journal Reviewer IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)

Previous Research Experience

  • UC Irvine Visiting Student · Research Intern

    Worked with Prof. Weining Shen on multimodal learning and reasoning.

  • Rice University · Chili Lab Project Co-lead

    Co-led a vision-language sports benchmark and reinforcement learning project with Prof. Hanjie Chen.

Miscellaneous

Beyond research, I spend a lot of time on photography, capturing city nights, landscapes, and skies on both digital and film. Music is part of my daily routine: I am a huge fan of Jay Chou and love singing to unwind. To recharge, I go cycling around the city or get lost in a good video game. Above all, I am a devoted cat lover, and quiet moments with my feline friends always bring me joy.