Metraipolis › archive › dc_1f2da044e16a

A Controlled Study of Models, Skills, Feedback, and Curriculum in Agentic Robotics — arXiv:2611.09030

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09030
License (as archived)
CC0-1.0
Archived
2026-11-09T22:34:45.712Z
SHA-256 of the archived text
bb1d4cb5609156326bdc4f01414856b52317c4dd11cc5f8abf72304d4a64a495
Stable URL of this page
https://metraipolis.com/documento/dc_1f2da044e16a

How to cite

A Controlled Study of Models, Skills, Feedback, and Curriculum in Agentic Robotics — arXiv:2611.09030 — arXiv (en), https://arxiv.org/abs/2611.09030 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1f2da044e16a (Metraipolis archive dc_1f2da044e16a, sha256 bb1d4cb5609156326bdc4f01414856b52317c4dd11cc5f8abf72304d4a64a495)
BibTeX
@misc{dc1f2da044e16a,
  title = {A Controlled Study of Models, Skills, Feedback, and Curriculum in Agentic Robotics — arXiv:2611.09030},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09030},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1f2da044e16a; sha256 bb1d4cb5609156326bdc4f01414856b52317c4dd11cc5f8abf72304d4a64a495},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09030. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agentic robotics asks coding agents to implement and iteratively improve robot policies from simulated and real-world physical interactions. We study the self-learning loop of agentic robotics through a controlled simulation study spanning nine manipulation tasks, 631 different experiment configurations, and 28K coding agent rollouts, systematically varying model capability, skill priors, feedback structure, and curriculum. The experiment data suggests that the bottleneck is often not the access to sufficient information, but interpreting it into the right policy update. Across these factors, a successful self-learning loop depends on how effectively agents attribute failures, reason task-relevant spatial and geometric structure, and selectively restructure code to improve policies.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.09030 — 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 2026-11-09T22:30:47.315Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).