Metraipolis › archive › dc_1dbe6da733ef

Video Prediction Policy 2: Predict Better, Act Better — arXiv:2610.10270

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10270
License (as archived)
CC0-1.0
Archived
2026-10-08T12:47:59.485Z
SHA-256 of the archived text
6191fdcc804b0b5145dc9a9240923cd9f5feadeeab709d7a24f4f1ceeaf79699
Stable URL of this page
https://metraipolis.com/documento/dc_1dbe6da733ef

How to cite

Video Prediction Policy 2: Predict Better, Act Better — arXiv:2610.10270 — arXiv (en), https://arxiv.org/abs/2610.10270 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1dbe6da733ef (Metraipolis archive dc_1dbe6da733ef, sha256 6191fdcc804b0b5145dc9a9240923cd9f5feadeeab709d7a24f4f1ceeaf79699)
BibTeX
@misc{dc1dbe6da733ef,
  title = {Video Prediction Policy 2: Predict Better, Act Better — arXiv:2610.10270},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10270},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1dbe6da733ef; sha256 6191fdcc804b0b5145dc9a9240923cd9f5feadeeab709d7a24f4f1ceeaf79699},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10270. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

World action models (WAMs) have emerged as an important class of generalist robot policies, aiming to transfer video prediction priors to action learning. However, we find that existing WAMs frequently produce incorrect motion predictions in open-ended environment, leading to erroneous actions. We attribute this limitation to two factors: (1) base video models are not optimized for manipulation, and (2) naively incorporating action components into video models can substantially degrade their generalization capabilities. We introduce Video Prediction Policy 2 (VPP2), a WAM that enables strong zero-shot generalization in both video prediction and action generation. First, we curate a large-scale, diverse dataset of manipulation videos to continue pretraining the base video foundation model. We annotate video clips with detailed captions and perform \textit{event-level} video pretraining to promote generalization across open-ended manipulation tasks. Second, we post-train and distill the video model into a single-step visual planner with fixed prediction horizon. Finally, we introduce action module via a mixture-of-transformers (MoT) architecture to learn implicit inverse dynamics model. Experiments demonstrate three key results: (1) VPP2-14B outperforms Cosmos3-64B by 11.0\% points in video prediction instruction-following success rate on open-ended tasks; (2) VPP2 surpasses the strongest baseline by 18.5\% points in success rate on real-world zero-shot ALOHA manipulation tasks; and (3) following benchmark-specific post-training, VPP2 achieves the highest success rates among evaluated methods on the challenging LIBERO-Pro, LIBERO-OOD, and RoboDojo benchmarks.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.10270 — 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 2026-10-08T12:18:15.469Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).