Sung-Wook Lee (이성욱)

I'm a second-year Computer Science PhD student and a member of the Link Lab at University of Virignia. I am currently advised by Professor Yen-Ling Kuo.

Before joining the PhD progam, I graduated from Virginia Tech with a bachelor's degree and a master's degree in Mechanical Engineering, where I was advised by Professor Alan Asbeck.

Email / CV / Scholar / Github

Research

My current interests are deep learning-based methods for robot manipulation, robot learning from humans, representation learning under data constraints, etc. Some papers are highlighted.

	TBD-VLA: Temporal Block Diffusion Vision Language Action Model Sung-Wook Lee, Xuhui Kang, Yen-Ling Kuo CoRL 2026 project page / arXiv Block diffusion meets Vision-Language-Action (VLA) models. TBD-VLA is a discrete, fast, and temporally-aware VLA model!
	CLASS: Contrastive Learning via Action Sequence Supervision for Robot Manipulation Sung-Wook Lee, Xuhui Kang, Brandon Yang, Yen-Ling Kuo CoRL 2025 project page / arXiv Aligning vision and action representations at a trajectory level induces useful invariant properties.
	Diff-DAgger: Uncertainty Estimation with Diffusion Policy for Robotic Manipulation Sung-Wook Lee, Xuhui Kang, Yen-Ling Kuo IEEE ICRA 2025 project page / arXiv Diffusion policy meets robot-gated DAgger (Data Aggregation).
	A Deep Learning-Based Approach for Foot Placement Prediction Sung-Wook Lee, Alan Asbeck IEEE RAL 2023 ieeexplore The GRU-based model architecture processes IMU sensor data and accurately predicts the next foot placement during human walking.

Website template cloned from here.