Assistant Professor, ShanghaiTech University
3 papers at NeurIPS 2025
This paper explores the representation of robot actions in the frequency domain and introduces FreqPolicy, a novel frequency-domain autoregressive framework for hierarchical robotic action generation.
Introduce the first Open-World Hand-Object Interaction (HOI) Synthesis framework that can generate Long-horizon HOI sequences of Unseen Objects from Open-vocabulary instruction with 3D Multimodal Large Language Model.
We propose GenPO, which effectively incorporates invertible diffusion model into on-policy RL, and deals with the challenge of log-likehood computation in diffusion policies.
