Yunzhong He
Papers
2
Total Citations
48
H-Index
2
About
Yunzhong He is a researcher at the intersection of robotics, natural language processing, and computer vision, with a primary focus on enabling more intuitive human-robot collaboration. His work centers on developing frameworks that allow robots to learn complex tasks by jointly interpreting language instructions and visual demonstrations. In his most-cited paper (42 citations), He introduced a novel approach using And-Or Graph (AoG) representations to ground hierarchical task structures from both linguistic and visual inputs, enabling cognitive robots to understand and execute multi-step procedures. This work bridges the gap between symbolic reasoning and perceptual learning, making it easier for non-experts to communicate with robots. He further extended this line of research by exploring how robots can learn human utility functions from video demonstrations, allowing for deductive planning that aligns with human preferences. His contributions are particularly valuable for advancing embodied AI systems that can operate in dynamic, real-world environments, and his citation impact reflects growing interest in multimodal learning for robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2