Metraipolis › archive › dc_bb40dbf71694

Towards VLA-Dreamer: Refining VLA Behavior Using World Models — arXiv:2609.31313

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31313
License (as archived)
CC0-1.0
Archived
2026-10-08T01:39:51.801Z
SHA-256 of the archived text
7b78c86ee6f0bb2fa9e3781bc846f8abc7fa0593e2cc66c9c1c32567cfb783eb
Stable URL of this page
https://metraipolis.com/documento/dc_bb40dbf71694

How to cite

Towards VLA-Dreamer: Refining VLA Behavior Using World Models — arXiv:2609.31313 — arXiv (en), https://arxiv.org/abs/2609.31313 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bb40dbf71694 (Metraipolis archive dc_bb40dbf71694, sha256 7b78c86ee6f0bb2fa9e3781bc846f8abc7fa0593e2cc66c9c1c32567cfb783eb)
BibTeX
@misc{dcbb40dbf71694,
  title = {Towards VLA-Dreamer: Refining VLA Behavior Using World Models — arXiv:2609.31313},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31313},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bb40dbf71694; sha256 7b78c86ee6f0bb2fa9e3781bc846f8abc7fa0593e2cc66c9c1c32567cfb783eb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31313. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-Language-Action models (VLAs), while showing strong potential for robot control, require massive amounts of high-quality imitation learning data. Moreover, the absence of an explicit world model casts further doubt on their control capabilities. In this concept paper, we propose a novel architecture that addresses sample efficiency in VLAs by training a predictive world model on the embedding space of the VLA's vision encoder. We hypothesize that these embeddings are action-relevant and usable for future prediction. To this end, we propose using the suggested architecture to investigate how well these embeddings predict the future based on actions, as the inability to do so would mark a key limitation of VLA architectures: the lack of a non-lossy implicit world model to simulate real-world dynamics. The proposed architecture differs from the standard world model dynamics as the loss comes from the embedding space rather than the pixel space, similar to joint embedding predictive architectures. Furthermore, the trained world model can be utilized for short-term planning tasks by sampling VLA actions given goal images. We intend to examine the richness of vision embeddings in VLAs and reduce their high data requirements through a world model that can also generate plans during inference.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.31313 — 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 2026-10-08T01:12:14.573Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).