Hosu Lee 이호수

KAIST · Integrated Vision & Language Lab (IVY Lab), advised by Prof. Yong Man Ro

leehosu01@kaist.ac.kr GitHub Google Scholar Hugging Face

Research Interests

Video understanding and multimodal large language models — long-form and streaming video comprehension, frame selection, and memory-efficient sequence modeling.

Publications

* denotes equal contribution.

PRISM: Predictive Recomposition via Semantic Latent Decomposition for View-invariant Video Representation Learning
Youngchae Chee*, Hosu Lee*, Sungjune Park, Junho Kim, Yong Man Ro
EMNLP 2026
STRIDE: When to Speak Meets Sequence Denoising for Streaming Video Understanding
Junho Kim*, Hosu Lee*, James M. Rehg, Minsu Kim, Yong Man Ro
ECCV 2026arXiv:2603.27593 · project page · code
ReFoCUS: Reinforcement-guided Frame Optimization for Contextual Understanding
Hosu Lee*, Junho Kim*, Hyunjun Kim, Yong Man Ro
CVPR 2026 Findingspp. 8291–8302 · arXiv:2506.01274 · project page · code
SALOVA: Segment-Augmented Long Video Assistant for Targeted Retrieval and Routing in Long-Form Video Analysis
Junho Kim*, Hyunjun Kim*, Hosu Lee, Yong Man Ro
CVPR 2025pp. 3352–3362 · arXiv:2411.16173 · project page · code
Look Every Frame All at Once: Video-Ma²mba for Efficient Long-form Video Understanding with Multi-Axis Gradient Checkpointing
Hosu Lee*, Junho Kim*, Hyunjun Kim, Yong Man Ro
Under reviewarXiv:2411.19460 · project page · code

Open Source