4 papers
KITE: Decoupling Kinematics and Interaction for Zero-Shot Cross-Embodiment Manipulation
Qianxu Wang, Kuan Fang
Generalizing manipulation policies across robot embodiments remains difficult because standard policies entangle task reasoning with embodiment-specific motor control. We study zer…
Unified Motion-Action Modeling for Heterogeneous Robot Learning
Yunhao Cao, Shitong Liu, Chao Feng +4
We present Unified Motion-Action (UMA) Model, an approach that uses 3D object motion trajectories as a shared interface to bridge visuomotor control and dynamics modeling. UMA trea…
Generate, Transfer, Adapt: Learning Functional Dexterous Grasping from a Single Human Demonstration
Xingyi He, Adhitya Polavaram, Yunhao Cao +4
Functional grasping with dexterous robotic hands is a key capability for enabling tool use and complex manipulation, yet progress has been constrained by two persistent bottlenecks…
Correspondence-Oriented Imitation Learning: Flexible Visuomotor Control with 3D Conditioning
Yunhao Cao, Zubin Bhaumik, Jessie Jia +2
We introduce Correspondence-Oriented Imitation Learning (COIL), a conditional policy learning framework for visuomotor control with a flexible task representation in 3D. At the cor…