Jingxiang Guo
Papers
3
Total Citations
8
H-Index
2
About
Jingxiang Guo is a rising roboticist whose research lies at the intersection of dexterous manipulation, task planning, and vision-language models. His work fundamentally addresses how robots can interact with complex, deformable, and everyday objects—from garments to furniture—with human-like precision. Guo’s most notable contribution is the **$\mathcal{D}(\mathcal{R}, \mathcal{O})$ Grasp** framework, which introduces a unified representation for robot-object interaction, enabling cross-embodiment dexterous grasping (4 citations). He further pushes the boundaries of autonomous assembly with **Manual2Skill**, a system that learns to read human instruction manuals to acquire furniture assembly skills using vision-language models (3 citations). In **MetaFold**, Guo tackles the notoriously difficult problem of garment folding by disentangling task planning from trajectory generation, leveraging foundation models for language-guided, multi-category folding (1 citation). Though early in his career, his work is distinguished by its focus on bridging high-level semantic understanding with low-level physical control. By integrating language models directly into manipulation pipelines, Guo is pioneering a new paradigm for robots that can understand instructions, adapt to novel objects, and perform complex, fine-grained tasks—a critical step toward truly capable home and industrial assistants.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3