Metraipolis › archive › dc_2b25ecad902f

RoboCode: Learning Reusable Action Primitives for Robot Manipulation — arXiv:2610.32225

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.32225
License (as archived)
CC0-1.0
Archived
2026-10-26T20:14:14.866Z
SHA-256 of the archived text
c7901bf201a6ab7d4bdbc65dc60b1259958982af22c0a1f80959c9ca204a7c4d
Stable URL of this page
https://metraipolis.com/documento/dc_2b25ecad902f

How to cite

RoboCode: Learning Reusable Action Primitives for Robot Manipulation — arXiv:2610.32225 — arXiv (en), https://arxiv.org/abs/2610.32225 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2b25ecad902f (Metraipolis archive dc_2b25ecad902f, sha256 c7901bf201a6ab7d4bdbc65dc60b1259958982af22c0a1f80959c9ca204a7c4d)
BibTeX
@misc{dc2b25ecad902f,
  title = {RoboCode: Learning Reusable Action Primitives for Robot Manipulation — arXiv:2610.32225},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.32225},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2b25ecad902f; sha256 c7901bf201a6ab7d4bdbc65dc60b1259958982af22c0a1f80959c9ca204a7c4d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32225. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent robot manipulation policies commonly predict continuous action chunks directly from observations, without explicitly modeling reusable structure in the action space. However, manipulation demonstrations often contain recurring short-horizon motion patterns that are shared across tasks. We introduce \textsc{RoboCode}, which learns these patterns through a shared soft-gated primitive action interface. Rather than assigning each action chunk to a single discrete code, \textsc{RoboCode} represents it as a soft composition of multiple learned primitives, providing a flexible intermediate representation of short-horizon actions. The primitive interface is learned from action chunks pooled across multiple tasks and then frozen while task-specific state encoders are trained to map current state inputs into the shared latent space. Experiments on five ManiSkill manipulation tasks address three questions. First, the learned primitive interface reconstructs action chunks with an average reconstruction MSE of 0.0187. Second, analysis of code activations reveals structured code reuse across tasks. Third, \textsc{RoboCode} achieves the highest or tied closed-loop success rate on all evaluated tasks, with an average success rate of 75.6%, compared with 68.8% for ACT and 41.6% for the larger Diffusion Policy baseline. These results suggest that soft-gated primitive codebooks provide an effective and interpretable intermediate action representation for robot manipulation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.32225 — 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 2026-10-26T19:49:03.473Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).