Chenyang Gu
Papers
5
Total Citations
22
H-Index
2
About
Chenyang Gu is an emerging researcher at the forefront of robotic manipulation and 3D visual perception, with work spanning foundation models for robotics, multi-embodiment learning, and spatial scene understanding. His most significant contribution is RoboMIND, a large-scale benchmark dataset comprising 107,000 demonstration trajectories across 479 diverse tasks and 96 object classes, collected through human teleoperation to advance generalized robot manipulation research — a work that has already garnered 14 citations since its 2025 release. Gu has also made notable strides in bridging 2D foundation models with 3D robotic reasoning through his Lift3D Policy framework, addressing the critical challenge of spatial perception in manipulation tasks. His Fast-in-Slow dual-system architecture tackles the tension between high-level reasoning and real-time execution efficiency in foundation policies, while his SliceOcc work advances indoor 3D semantic occupancy prediction using innovative vertical slice representations. Collectively, Gu's research reflects a coherent vision: building robust, generalizable robotic systems that can perceive, reason, and act effectively in complex real-world environments — positioning him as a promising contributor to the next generation of embodied AI research.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5