Metraipolis › archive › dc_87f70963676b

IronMan: Information-Constrained Video-Action Learning for Robot Manipulation — arXiv:2610.07961

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07961
License (as archived)
CC0-1.0
Archived
2026-10-07T13:12:12.572Z
SHA-256 of the archived text
da5e88904de4d0ffdc21e0dee0aa17d203ef8e95a6f83c938211f8813d0df98c
Stable URL of this page
https://metraipolis.com/documento/dc_87f70963676b

How to cite

IronMan: Information-Constrained Video-Action Learning for Robot Manipulation — arXiv:2610.07961 — arXiv (en), https://arxiv.org/abs/2610.07961 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_87f70963676b (Metraipolis archive dc_87f70963676b, sha256 da5e88904de4d0ffdc21e0dee0aa17d203ef8e95a6f83c938211f8813d0df98c)
BibTeX
@misc{dc87f70963676b,
  title = {IronMan: Information-Constrained Video-Action Learning for Robot Manipulation — arXiv:2610.07961},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07961},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_87f70963676b; sha256 da5e88904de4d0ffdc21e0dee0aa17d203ef8e95a6f83c938211f8813d0df98c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07961. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Video Action Models (VAMs) couple visual dynamics modeling with action generation for robot manipulation. However, video representations are not naturally suited to action generation, as exposing the action policy to excessive visual detail can impair its generalization ability. Therefore, we introduce IronMan (Information-constRained videO-actioN learning for robot MANipulation), a robust video-action learning framework built on the information bottleneck principle. The core principle of this framework is to impose information constraints that suppress irrelevant visual information while preserving action-relevant dynamics cues. IronMan employs a dynamics-aware bottleneck that distills noisy, entangled one-step video features into compact world representations. Extensive simulation and real-world experiments demonstrate strong in-distribution (ID) performance and out-of-distribution (OOD) robustness while maintaining efficient inference. IronMan achieves success rates of 99.0% on LIBERO and 79.4% on RoboTwin clean2clean, outperforming all the evaluated baselines. Under OOD shifts, IronMan achieves a success rate of 79.1% on LIBERO-Plus, exceeding the strongest baseline by 10.4 percentage points. Project page: this https URL

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/2610.07961 — 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 2026-10-07T12:42:33.590Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).