Metraipolis › archive › dc_2103acf25dd5
Devol-ONE: One Autoregressive Mixture of Transformers to Unify Vision-Language-Action and Latent World Modeling — arXiv:2609.32193
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32193
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:50:46.629Z
- SHA-256 of the archived text
d6bdc8d123a372a8a9082e139fc6188f070c25b3899e8f61b6a3d5aaa87bbb4d- Stable URL of this page
https://metraipolis.com/documento/dc_2103acf25dd5
How to cite
Devol-ONE: One Autoregressive Mixture of Transformers to Unify Vision-Language-Action and Latent World Modeling — arXiv:2609.32193 — arXiv (en), https://arxiv.org/abs/2609.32193 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2103acf25dd5 (Metraipolis archive dc_2103acf25dd5, sha256 d6bdc8d123a372a8a9082e139fc6188f070c25b3899e8f61b6a3d5aaa87bbb4d)
BibTeX
@misc{dc2103acf25dd5,
title = {Devol-ONE: One Autoregressive Mixture of Transformers to Unify Vision-Language-Action and Latent World Modeling — arXiv:2609.32193},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32193},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2103acf25dd5; sha256 d6bdc8d123a372a8a9082e139fc6188f070c25b3899e8f61b6a3d5aaa87bbb4d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32193. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision Language Action (VLA) models condition actions directly on current visual and language context, without an explicit account of how the scene evolves under candidate actions. World Action Models (WAM) attempt to address this limitation by predicting future states, but existing designs keep prediction and policy learning architecturally separate, connecting them only through the predicted output, whether through pixel space video generation or a latent forecasting module trained independently of the policy. We present Devol-ONE, a Mixture of Transformers architecture that unifies vision language understanding, latent world dynamics prediction, and action generation within a single autoregressive framework. Instead of encoding vision language tokens once and feeding them to the action expert, Devol-ONE runs autoregressive prediction jointly across a vision language stream and a V-JEPA pretrained dynamics stream, attending to the vision language key-value cache at every layer to forecast future latent states under language guidance. The action expert is in turn shaped continuously by semantic reasoning and predicted physical dynamics rather than by a fixed representation computed in advance. Extensive experiments are conducted on LIBERO, LIBERO-PLUS, RoboTwin2.0 along with real-world evaluation on Flexiv single-arm and dual-arm setups. Ablation studies show the effectiveness of dynamic stream prediction and layer-wise unified attention to validate our model architectural coherency. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.32193 — 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 2026-10-07T23:29:26.982Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).