Metraipolis › archive › dc_d22a161d010c

ODEAction: From Predicting Chunks to Evolving Actions — arXiv:2611.09672

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09672
License (as archived)
CC0-1.0
Archived
2026-11-09T22:46:07.832Z
SHA-256 of the archived text
cfd6562bf0e79441f231024315cdf4267484b657e3deb0014af08ccc935b2aa8
Stable URL of this page
https://metraipolis.com/documento/dc_d22a161d010c

How to cite

ODEAction: From Predicting Chunks to Evolving Actions — arXiv:2611.09672 — arXiv (en), https://arxiv.org/abs/2611.09672 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d22a161d010c (Metraipolis archive dc_d22a161d010c, sha256 cfd6562bf0e79441f231024315cdf4267484b657e3deb0014af08ccc935b2aa8)
BibTeX
@misc{dcd22a161d010c,
  title = {ODEAction: From Predicting Chunks to Evolving Actions — arXiv:2611.09672},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09672},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d22a161d010c; sha256 cfd6562bf0e79441f231024315cdf4267484b657e3deb0014af08ccc935b2aa8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09672. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-action (VLA) models typically generate actions in chunks, organizing continuous robot execution into separately predicted segments and thus introducing potential discontinuities at action-chunk transitions. We propose ODEAction, a chunk-free action generation paradigm that replaces action-chunk prediction with Continuous Action Evolution. Given visual observations, language instructions, and robot proprioceptive information, ODEAction learns an action transition field defined over action space, where a transition vector characterizes the evolution rate of the current action with respect to evolution time. During execution, the model takes the current action as its initial condition and integrates the learned field using an ordinary differential equation (ODE), recursively generating subsequent actions with explicit dependencies. ODEAction is implemented as an action head and does not depend on a specific vision-language backbone architecture. Experiments on LIBERO, EBench, RoboTwin, and real robots show that ODEAction improves both task performance and trajectory quality over matched action-chunking baselines, with particularly pronounced gains on fine manipulation tasks.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2611.09672 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-09T22:38:02.204Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).