Mingdong Wu
Papers
1
Total Citations
7
H-Index
1
About
Mingdong Wu is a rising researcher in embodied AI and robotic manipulation, with a focus on functional object rearrangement—a core challenge in enabling robots to interact intelligently with unstructured environments. His work addresses how robots can learn to rearrange objects not merely by position, but by understanding and satisfying functional goals, such as placing a mug on a hook or a book on a shelf. His most cited paper, "LVDiffusor: Distilling Functional Rearrangement Priors From Large Models Into Diffusor" (2024), introduces a novel framework that distills functional priors from large pre-trained models into a diffusion-based policy, enabling robots to generalize rearrangement strategies across diverse objects and configurations. This work has already garnered 7 citations within its first year, signaling strong early impact. Wu’s research bridges the gap between high-level semantic understanding and low-level robotic control, offering a scalable path toward more adaptable and task-aware robots. His contributions are particularly valuable for students and researchers interested in the intersection of generative models, robot learning, and functional reasoning in physical spaces.
Research Focus
Key Achievements
Top Papers
- 1