Metraipolis › archive › dc_02ade0d10dee

arXiv 2610.10515: RoboJEPA: Scaling Robotic Latent World Models

Source: Metraipolis archive
Original: https://arxiv.org/abs/2610.10515
License (as archived): CC-BY-4.0
Archived: 2026-10-08T03:51:11.685Z
SHA-256 of the archived text: d00296c90f85be6877904c89a7606dfc342356dbba2228d28ef210302f7a89e1
Stable URL of this page: https://metraipolis.com/documento/dc_02ade0d10dee

How to cite

arXiv 2610.10515: RoboJEPA: Scaling Robotic Latent World Models — Metraipolis archive, https://arxiv.org/abs/2610.10515 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_02ade0d10dee (Metraipolis archive dc_02ade0d10dee, sha256 d00296c90f85be6877904c89a7606dfc342356dbba2228d28ef210302f7a89e1)

BibTeX

@misc{dc02ade0d10dee,
  title = {arXiv 2610.10515: RoboJEPA: Scaling Robotic Latent World Models},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.10515},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_02ade0d10dee; sha256 d00296c90f85be6877904c89a7606dfc342356dbba2228d28ef210302f7a89e1},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.10515. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Latent world models have shown a remarkable ability to predict future states and to plan in the real world. In practice, however, we lack a principled way to estimate how their capabilities scale with model size, data, and compute, an open problem that slows progress in the field. In this work we present RoboJEPA, a world model based on the Joint Embedding Predictive Architecture (JEPA) and trained on a large-scale dataset spanning 12 robotic embodiments. We show that RoboJEPA's imagination error, the error of its latent rollouts, follows a second-order power law in compute, allowing us to predict model quality well beyond the scale at which the law is fit. We further show that downstream robotic planning performance improves predictably with compute, and that imagination error is strongly correlated with it, making it a reliable proxy for real-robot evaluation. Finally, we demonstrate that latent world models can be deployed zero-shot as robotic agents, planning toward a single goal image to solve tasks requiring long-horizon planning on real hardware. We release all model checkpoints together with our training and robot deployment code. To our knowledge, this is the first work to establish scaling laws for multi-embodiment robotic world models trained on real robot data, and RoboJEPA, at 8B parameters, is the largest JEPA predictor model trained to date.

Articolo: «RoboJEPA: Scaling Robotic Latent World Models» — Zholus, Artem; Beltran-Velez, Nicolas; Yuan, Jianhao; Chandar, Sarath; Nagarajan, Tushar et al. (2026-10-07), arXiv:2610.10515, categoria cs.AI (intelligenza artificiale).
Fonte: https://arxiv.org/abs/2610.10515
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-08T03:51:01Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.