ODE Policy learns action dynamics via neural ODEs for multimodal robot trajectories
ntumars · hf · 2026-11-10
ODE Policy learns action dynamics rather than actions themselves using neural ODEs, generating continuous, multimodal robot trajectories.
It improves imitation and reinforcement learning across diverse tasks — unimanual/bimanual manipulation, dexterous hands, loco-manipulation, and locomotion — working with Visuomotor, VLA, WAM, and RL actor backbones.
More from Embodied
- Day 5 of a DIY robot arm: CAD render of the differential wrist in progress — _Stocko_ · 2026-11-10
- SuperMimic's 30× gradient SNR edge over score-function cuts imagined steps 64× vs PPO — daibond_alpha · 2026-11-10
- One frozen controller for humanoids: spectral skills enable motion composition and language control — daibond_alpha · 2026-11-10
- SuperMimic trains humanoid tracker in 3 hours on one GPU, beating 128-GPU SoTA baseline — daibond_alpha · 2026-11-10
- Johns Hopkins lab hiring postdocs in embodied intelligence, applications due Dec 11 — mangahomanga · 2026-11-10
- HARP-Bench: learned robot policies score 6-19% on 37 tasks involving humans — gan_chuang · 2026-11-10