Hi! My name is Sung-hwan Han. I am an M.S. student in the High Performance AI System Lab at Sogang University, advised by Prof. Youngmin Yi.
My research focuses on two areas:
- Efficient inference and deployment of AI models, particularly Vision-Language-Action (VLA) models
- Human-Robot Interaction (HRI), with an emphasis on Robots for Good
Across both areas, I am driven by a single goal: developing AI that benefits people who are often overlooked or underserved. I hope my research helps make intelligent systems more accessible, practical, and meaningful for those who need them most.
news
| Jun 17, 2026 | One paper was accepted to IROS 2026. |
|---|---|
| May 31, 2026 | One paper was accepted to AdaptFM Workshop @ ICML 2026. [Link] |
selected publications
- IROS 2026
AdaVLA: Adaptive Step Flow Matching for Training-Free Acceleration of Vision-Language-Action ModelsIn IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2026To appear - AdaptFM
Zero-Shot Quantization for Vision-Language-Action Models via Trajectory Curvature and Attention GuidanceIn AdaptFM: Resource-Adaptive Foundation Model Inference, 2026