Metraipolis › archive › dc_2ea77d9df324

Unifying Embodied Foundation Models via a Model-to-Policy Formulation — arXiv:2610.23053

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.23053
License (as archived): CC0-1.0
Archived: 2026-10-19T08:43:07.329Z
SHA-256 of the archived text: 43001f47155223a0ca58e0903042cda549392527ee4c1cbce38076b1fda8bf77
Stable URL of this page: https://metraipolis.com/documento/dc_2ea77d9df324

How to cite

Unifying Embodied Foundation Models via a Model-to-Policy Formulation — arXiv:2610.23053 — arXiv (en), https://arxiv.org/abs/2610.23053 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2ea77d9df324 (Metraipolis archive dc_2ea77d9df324, sha256 43001f47155223a0ca58e0903042cda549392527ee4c1cbce38076b1fda8bf77)

BibTeX

@misc{dc2ea77d9df324,
  title = {Unifying Embodied Foundation Models via a Model-to-Policy Formulation — arXiv:2610.23053},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23053},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2ea77d9df324; sha256 43001f47155223a0ca58e0903042cda549392527ee4c1cbce38076b1fda8bf77},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23053. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-action (VLA) models and world-action models (WAMs) have emerged as promising approaches to robot learning and control. These models integrate multimodal perception and action generation, but differ in their architectures, output representations, and learning objectives. In this paper, we present a unified, architecture-agnostic probabilistic formulation for embodied foundation models based on stochastic kernels. Our formulation has two key components: (i) we characterize embodied foundation models through conditional output distributions and action-realization kernels, which together induce task-conditioned robot policies; and (ii) we connect pretraining and post-training to policy deployment, accommodating model-specific pretraining objectives and offline or online post-training strategies. We instantiate the formulation for OpenVLA and DreamZero, covering autoregressive action-token prediction and joint video-action generation.

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/2610.23053 — 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 2026-10-19T08:39:10.140Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).