Hi there! I am a senior undergraduate from Shanghai Jiao Tong University. Currently, I am a robotics researcher at Machine Vision and Intelligence Group (supervised by Prof. Cewu Lu). Besides, I’m a member of Guozhi class (held by Prof. Xiaoou Tang). Previously, I worked as a research intern at OpenRobotLab (supervised by Dr. Jiangmiao Pang).
My research interest lies in Robot manipulation and Embodied AI. During off time, I am a music enthusiast, fond of a wide range of genres including Hip-Hop, R&B, Rock, Pop, and Electronic. Sports take up much of my leisure as well, among which basketball is my favourite. I am also a fan of travelling and cinematography.
🔥 News
- 2025.04: 🎉🎉 One paper (Novel Demonstration Generation with Gaussian Splatting Enables Robust One-Shot Manipulation) is accepted by RSS 2025!
- 2024.09: 🎉🎉 One paper (Learning H-Infinity Locomotion Control) is accepted by CoRL 2024!
- 2023.08: Joined OpenRobotLab.
- 2023.02: Joined Machine Vision and Intelligence Group.
📝 Selected Publications
Novel Demonstration Generation with Gaussian Splatting Enables Robust One-Shot Manipulation
Sizhe Yang*, Wenye Yu*, Jia Zeng, Jun Lv, Kerui Ren, Cewu Lu, Dahua Lin, Jiangmiao Pang†
[Project website] [Paper] [Code]
RoboSplat is framework that leverages 3D Gaussian Splatting (3DGS) to generate novel demonstrations for RGB-based policy learning in a one-shot manner, enabling robust performance across six types of visual generalization in the real world.
Learning H-Infinity Locomotion Control 🎉 Best Poster Award at CoRL 2024 LocoLearn Workshop
Junfeng Long*, Wenye Yu*, Quanyi Li*, Zirui Wang, Dahua Lin, Jiangmiao Pang†
We propose H-Infinity Locomotion Control, an adversarial framework for quadrupedal robots which enhances their robustness against external forces with a performance guarantee.
📝 Other Publications
SIME: Enhancing Policy Self-Improvement with Modal-level Exploration
Yang Jin*, Jun Lv*, Wenye Yu, Hongjie Fang, Yong-Lu Li, Cewu Lu†
[Project website] [Paper] [Code]
With modal-level exploration, the robot can generate more diverse and multi-modal interaction data. By learning from the most valuable trials and high-quality segments from these interactions, the robot can effectively refine its capabilities through self-improvement.
🎖 Honors and Awards
- 2024.06 SenseTime Scholarship (25/year in China)
- 2023.09 National Scholarship
- 2022.09 National Scholarship (Highest Honor for undergraduates in China)
📖 Educations
- 2021.09 - now, Shanghai Jiao Tong University
- 2018.09 - 2021.06, Shanghai High School
- TOEFL score: 110 [Listening: 28, Reading: 30, Speaking: 25, Writing: 27] (first time)
- CET6 score: 677 (first time)
💻 Experiences
- 2023.02 - now, Machine Vision and Intelligence Group, Shanghai Jiao Tong University.
- 2023.08 - 2025.04, OpenRobotLab, Shanghai AI Laboratory.