Scholar
Xuanlin Li
Google Scholar ID: 7vyVxxQAAAAJ
Unknown affiliation
Citations & Impact
Citations
2,499
H-index
14
i10-index
15
Publications
19
Co-authors
20
list available
Publications
Resume (English only)
Academic Achievements
- Published papers include 'Planning-Guided Diffusion Policy Learning for Generalizable Contact-Rich Bimanual Manipulation', 'Evaluating Real-World Robot Manipulation Policies in Simulation', and more, presented at conferences such as CoRL 2024, ICRA 2024, NeurIPS 2023, etc.
Research Experience
- During his PhD, he has been a major contributor of the SAPIEN Manipulation Skill Challenge (ManiSkill). He also led the benchmark on evaluating real-world generalist robot manipulation policies in simulation (Simpler-Env).
Education
- PhD, UCSD CSE, advised by Prof. Hao Su, 2021-2025; B.A. Mathematics & B.A. Computer Science, UC Berkeley, 2017-2021; research assistant at Berkeley Artificial Intelligence Research, advised by Prof. Trevor Darrell.
Background
- Primary interests: Embodied AI, Vision-Language, and Robotics. Goal is to build robotic agents with universal, open-world manipulation, perception, and reasoning capabilities that can be efficiently and robustly deployed for real world applications. This is achieved by scaling up high-quality training data, RL & learning-from-demonstration algorithms, vision-language models, and evaluation benchmarks.
Miscellany
- Currently a research scientist and founding team member at Hillbot, Inc., focusing on building generalizable algorithms & systems for robotics.
