4 papers
RoboFFT: Finetuning generative robot policy via online reinforcement learning with forward process
Yu Li, Shenghe Hu, Yuhan Wang +4
Generative models, such as diffusion and flow-based models, have shown strong promise for robot policy learning by capturing complex and multimodal action distributions from demons…
Frequency-Aware Flow Matching for Continuous and Consistent Robotic Action Generation
Jianing Guo, Fangzheng Chen, Zihao Mao +12
Flow matching has emerged as a standard paradigm for robotic manipulation owing to its strong expressive power for modelling complex, multimodal action distributions, alongside sim…
Dexterous Non-Prehensile Manipulation for Ungraspable Object via Extrinsic Dexterity
Yuhan Wang, Yu Li, Yaodong Yang +1
Objects with large base areas become ungraspable when they exceed the end-effector's maximum aperture. Existing approaches address this limitation through extrinsic dexterity, whic…
RetrDex: Efficient Object Retrieval in Cluttered Scenes with a Dexterous Hand
Fengshuo Bai, Yu Li, Jie Chu +5
Retrieving objects buried beneath clutter is both challenging and time-consuming, as complex support relationships make manipulation particularly difficult. Existing methods either…