6 papers
PFM-HR: Pose Flow Matching for Humanoid Robots
Yukang Gao, Yi Gu, Yangchen Zhou +9
Motion priors improve reinforcement learning for physics-based humanoid tracking, but temporal priors require ordered motion clips, while pose priors provide limited guidance for p…
Try Once, Then Optimal: De-Redundified Procedure Memory for Cross-Episode Exploration Amortization
Haizhou Ge, Haochen Ouyang, Zhixing Chen +6
Manipulating objects with hidden internal state, such as a latched microwave, forces a robot to probe before it can act. Yet a robot that has solved an instance once re-runs the sa…
OmniContact: Chaining Meta-Skills via Contact Flow for Generalizable Humanoid Loco-Manipulation
Runyi Yu, Xiaoyi Lin, Ji Ma +11
Learning long-horizon humanoid loco-manipulation poses a dual challenge: it requires not only the robust execution of meta-skills but also their seamless, closed-loop chaining equi…
When Video Misreads: Closed-Loop Distillation of Reading Heuristics for Exploratory Manipulation Trace QA
Haizhou Ge, Yufei Jia, Yue Li +5
Exploratory manipulation often turns an apparent failed attempt into the key evidence for what to do next. For example, a robot pulls a locked cabinet drawer, fails, and only succe…
3DFlowAction: Learning Cross-Embodiment Manipulation from 3D Flow World Model
Hongyan Zhi, Peihao Chen, Siyuan Zhou +4
Manipulation has long been a challenging task for robots, while humans can effortlessly perform complex interactions with objects, such as hanging a cup on the mug rack. A key reas…
Learning 3D Persistent Embodied World Models
Siyuan Zhou, Yilun Du, Yuncong Yang +4
The ability to simulate the effects of future actions on the world is a crucial ability of intelligent embodied agents, enabling agents to anticipate the effects of their actions a…