Metraipolis › archive › dc_ea51f3973584

You Don't Need To Stay in The Loop: An Agentic Robotics Loop for Robot-Policy Improvement — arXiv:2608.07555

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.07555
License (as archived)
CC0-1.0
Archived
2026-10-19T09:12:20.734Z
SHA-256 of the archived text
2de7916235b453bffc53cfcf9339e7edae4921d3c538174788d27f61e4f3110e
Stable URL of this page
https://metraipolis.com/documento/dc_ea51f3973584

How to cite

You Don't Need To Stay in The Loop: An Agentic Robotics Loop for Robot-Policy Improvement — arXiv:2608.07555 — arXiv (en), https://arxiv.org/abs/2608.07555 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea51f3973584 (Metraipolis archive dc_ea51f3973584, sha256 2de7916235b453bffc53cfcf9339e7edae4921d3c538174788d27f61e4f3110e)
BibTeX
@misc{dcea51f3973584,
  title = {You Don't Need To Stay in The Loop: An Agentic Robotics Loop for Robot-Policy Improvement — arXiv:2608.07555},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.07555},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea51f3973584; sha256 2de7916235b453bffc53cfcf9339e7edae4921d3c538174788d27f61e4f3110e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.07555. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Coding agents such as Claude Code and Codex close the software loop: a main agent manages the loop, subagents analyze and execute, tools do the work. We port this architecture to robot-policy improvement, where one difference dominates the design: robotic tools---trained policies, training pipelines, data collection---fail routinely, so a tool's quality must be measured, recorded at every call, and expired when the artifact behind it changes. AgenticRobotics is a backend-independent control plane in which an LLM controller drives disposable workers through durable train--evaluate--improve transactions: an immutable objective, controller-owned measurement, commit-keyed crash recovery, an evidence-graded skill library, and a tool registry with a standardized, recorded call surface. The title is an operational claim, not a selection claim: the operator can leave because promotion is evidence-gated, state is recoverable, and capability quality is derived from records---not because the loop picks better checkpoints than a human; on the one lineage we measured, it does not. The gates measurably buy false-promotion control (0.001 per run hardened versus 0.005--0.021 shipped), anytime-valid decisions under optional stopping, zero lost or duplicate effects under kill injection, and six of six artifact-tampering classes caught by a signed verifier.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-11 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2608.07555 — 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 2026-10-19T09:11:58.714Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).