Papers
17
Total Citations
289
H-Index
9
About
Youngwoon Lee is a leading researcher in robot manipulation, with a focus on enabling robots to perform complex, long-horizon tasks through skill learning, imitation, and scalable data collection. His most impactful contribution is the DROID dataset (2024, 108 citations), a large-scale, in-the-wild manipulation dataset that has become a foundational resource for the field. Lee also pioneered the IKEA Furniture Assembly Environment (2021), one of the first benchmarks for long-horizon, hierarchical manipulation, and later introduced FurnitureBench (2023, 25 citations), a reproducible real-world benchmark designed to advance dexterous control and visual perception. His work on skill coordination (2020, 23 citations) and learned skill priors (2020, 19 citations) has advanced reinforcement learning by enabling agents to leverage prior experience and decompose complex tasks into reusable sub-skills. Lee also developed PATO (2023, 14 citations), a policy-assisted teleoperation system for scalable robot data collection, and explored multimodal learning from vision and touch (2024, 10 citations). With a consistent focus on bridging simulation and reality, Lee’s research provides critical tools and frameworks for building generalizable, data-driven robot manipulation systems.
Research Focus
Key Achievements
Top Papers
- 1DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset108 citations · 2024
- 2
- 3
- 4
- 5Accelerating Reinforcement Learning with Learned Skill Priors19 citations · 2020
- 6
- 7PATO: Policy Assisted TeleOperation for Scalable Robot Data Collection14 citations · 2023
- 8
- 9Demonstration-Guided Reinforcement Learning with Learned Skills9 citations · 2021
- 10