Jusuk Lee

Hello! I am a first year Ph.D. student under the supervision of Prof. H. Jin Kim at Seoul National University.

I am interested in the intersection of robotics, machine learning, and deep learning. I am currently focusing on learning robot skills from human videos and on multimodal robot learning beyond vision.

I am also fortunate to work with Prof. Furong Huang and Prof. Jia-Bin Huang at UMD.

News

Sep 2026: Two papers (DynaFLIP and FLAG) accepted to NeurIPS 2026!

Sep 2026: μ₀ accepted to CoRL 2026!

Jul 2026: BooST accepted to IEEE RA-L!

Mar 2024: I joined LARR at Seoul National University.

Publications

Dex-One2Many: Learning Dexterous Manipulation from a Single Human Demonstration

Jusuk Lee^*, Sungha Kim^*, Yeonsoo Park^*, Jonguk Cheon, Yoonkyo Jung, Yongjun You, H. Jin Kim, Jia-Bin Huang, Furong Huang, Youngseok Jang^†, Seungjae Lee^†

^* equal contribution ^† equal advising

Under review

project page arXiv github (coming soon)

μ₀: A Scalable 3D Interaction-Trace World Model

Seungjae Lee^*, Yoonkyo Jung^*, Jusuk Lee, Jonghun Shin, Amir Hossein Shahidzadeh, Yao-Chih Lee, H. Jin Kim, Jia-Bin Huang^†, Furong Huang^†

^* equal contribution ^† equal advising

CoRL 2026

project page arXiv github 🤗 hugging face

FLAG: Flow Policy MaxEnt-RL by Latent Augmented Guidance

Sungha Kim^*, Gawon Lee^*, Jusuk Lee, Jonghae Park, H. Jin Kim, Daesol Cho

^* equal contribution

NeurIPS 2026

project page arXiv github

Periodic Skill Discovery

Jonghae Park, Daesol Cho, Jusuk Lee, Dongseok Shim, Inkyu Jang, H. Jin Kim

NeurIPS 2025

project page arXiv github

Projects

Representative projects are highlighted.

Learning Cooperation with Learned Skills

Multi-agent reinforcement learning(MARL) typically requires the design of a sophisticated reward function to effectively guide multiple agents in learning desirable behaviors. However, this process is time-consuming and demands domain knowledge. In this paper, we train each agent to learn a diverse set of skills in advance. Then the agents are able to solve complex tasks by effectively coordinating these skills.
Mar. 2024 - Jun. 2024

paper github

Autonomous drone system

This project focuses on developing an autonomous drone system capable of real-time obstacle avoidance, building exploration, payload delivery to a specified location, and safe return to home base. The system has been successfully applied in both simulation and on real hardware.
Mar. 2023 - Sep. 2023

video github(real) github(simulation)

Education