About Me
Hi! I am Yuanhao Zou, a Computer Science PhD student at the University of Central Florida, advised by Prof. Chen Chen at the Center for Research in Computer Vision (CRCV).
I received my Master’s degree in Electrical and Computer Engineering from the University of Michigan, Ann Arbor, where I was advised by Prof. Zhaozheng Yin on Medical Vision and Language Models. Before that, I obtained my Bachelor’s degree in Computer Science from Central South University, working with Prof. Xiangjian He on Medical Image Segmentation.
I am currently an AI Research Intern at OpusClip, training the core Opus Clip Video Editing Agent with a focus on reinforcement-learning-based reasoning optimization.
Research Interests
My research focuses on building efficient and scalable multimodal models that can understand the visual world over time. Specifically, I am interested in:
- Video Understanding — Temporal grounding, long-form video reasoning, and frame selection.
- Efficient Vision-Language Models — Compact, on-device VLMs that keep strong perception and reasoning under tight compute budgets.
- Multimodal Agents and Reinforcement Learning — Joint reasoning across RGB, thermal, depth, and multi-view video.
- Video Anomaly Detection — Detecting and localizing anomalous events in real-world surveillance and edge scenarios.
Previously, I also worked extensively on Medical Vision-Language Models (visual question answering, image–text retrieval) and Medical Image Segmentation.
News
- [Oct 2026] Joined OpusClip as an AI Research Intern, training the core Opus Clip Video Editing Agent with RL-based reasoning optimization.
- [Oct 2026] CoMET-Bench / CoMET-Agent is accepted to NeurIPS 2026, Evaluations & Datasets Track!
- [Jun 2026] CoMET Agent and CoMET-Bench are released on arXiv. 🎉
- [Jun 2026] CLARITY is accepted to ECCV 2026!
- [Feb 2026] How Should Video LLMs Output Time? is accepted to a CVPR 2026 Workshop.
- [Jan 2026] A.I.R. is accepted to ICLR 2026!
- [Aug 2025] Started my PhD journey at UCF CRCV with Prof. Chen Chen. 🐊
Contact
Feel free to reach out via yuanhaoz@ucf.edu if you would like to chat about research or collaboration.