Honghui Dong
Papers
1
Total Citations
4
H-Index
1
About
Honghui Dong is a researcher at the forefront of embodied AI and robotic perception, with a particular focus on bridging the gap between high-level visual understanding and low-level physical constraints. His most cited work, "PhysVLM: Enabling Visual Language Models to Understand Robotic Physical Reachability" (2025, 4 citations), addresses a critical limitation in state-of-the-art vision-language models (VLMs). While VLMs excel at environmental perception, Dong identifies that they frequently generate inaccurate or impractical responses in embodied tasks because they lack an intrinsic understanding of a robot's physical reachability. His major contribution lies in designing PhysVLM to integrate geometric and kinematic reasoning directly into the visual-language pipeline, enabling models to not only "see" the world but also to infer what a robot can physically interact with. This work has immediate implications for safer and more reliable autonomous manipulation. Though early in its citation impact, Dong’s research is poised to influence how future VLMs are trained for real-world robotics, making his work essential reading for students and researchers interested in the intersection of computer vision, natural language processing, and robotic control.
Research Focus
Key Achievements
Top Papers
- 1