Metraipolis › archive › dc_5bfe59fe4e8d

MAGMA-GEN: Validated Recovery Supervision from Ambiguous Failures via Counterfactual Re-Execution — arXiv:2609.20056

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.20056
License (as archived): CC0-1.0
Archived: 2026-10-10T03:21:24.009Z
SHA-256 of the archived text: 043a102a568a612e54f47e9e648aeb1dfc3b9aab75ba01146e1f6dcc5f4c41d0
Stable URL of this page: https://metraipolis.com/documento/dc_5bfe59fe4e8d

How to cite

MAGMA-GEN: Validated Recovery Supervision from Ambiguous Failures via Counterfactual Re-Execution — arXiv:2609.20056 — arXiv (en), https://arxiv.org/abs/2609.20056 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5bfe59fe4e8d (Metraipolis archive dc_5bfe59fe4e8d, sha256 043a102a568a612e54f47e9e648aeb1dfc3b9aab75ba01146e1f6dcc5f4c41d0)

BibTeX

@misc{dc5bfe59fe4e8d,
  title = {MAGMA-GEN: Validated Recovery Supervision from Ambiguous Failures via Counterfactual Re-Execution — arXiv:2609.20056},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20056},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5bfe59fe4e8d; sha256 043a102a568a612e54f47e9e648aeb1dfc3b9aab75ba01146e1f6dcc5f4c41d0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20056. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hierarchical robotic systems executing long-horizon manipulation tasks must make high-level semantic decisions that orchestrate stochastic low-level skills. In this setting, failed rollouts are ambiguous: a poor downstream state may reflect an invalid high-level decision, partial observation, or a valid decision whose physical execution failed. Traditional supervised learning lacks data for such recovery states, while reinforcement learning struggles with sparse rewards and non-local credit assignment. We propose MAGMA-GEN, an on-policy data-generation pipeline that converts ambiguous failed rollouts into validated recovery supervision. MAGMA-GEN first uses a privileged coach to hypothesize an early decision-level error and propose localized correction or recovery actions. Because this diagnosis is fallible, candidates are retained only if re-execution from the same state under matched conditions improves downstream progress. This produces supervised examples from the agent's own failure distribution without per-step human demonstrations. Evaluated on interactive long-horizon manipulation tasks, MAGMA-GEN improves task success and recovery capabilities, against distillation and trajectory-repair baselines under evolving task constraints in both simulation and real-robot execution.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.20056 — 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 2026-10-10T03:04:13.692Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).