Metraipolis › archive › dc_a0643ccb1fa8

VR-JEPA: Learning Contrastive-State Latent Guidance for Generation-based Video Reasoning — arXiv:2609.40129

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.40129
License (as archived)
CC0-1.0
Archived
2026-10-07T14:46:36.761Z
SHA-256 of the archived text
eb3453b184622455b505ce6133ca6a17990bce791b5d002a953d1f87b2ea2bc5
Stable URL of this page
https://metraipolis.com/documento/dc_a0643ccb1fa8

How to cite

VR-JEPA: Learning Contrastive-State Latent Guidance for Generation-based Video Reasoning — arXiv:2609.40129 — arXiv (en), https://arxiv.org/abs/2609.40129 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a0643ccb1fa8 (Metraipolis archive dc_a0643ccb1fa8, sha256 eb3453b184622455b505ce6133ca6a17990bce791b5d002a953d1f87b2ea2bc5)
BibTeX
@misc{dca0643ccb1fa8,
  title = {VR-JEPA: Learning Contrastive-State Latent Guidance for Generation-based Video Reasoning — arXiv:2609.40129},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.40129},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a0643ccb1fa8; sha256 eb3453b184622455b505ce6133ca6a17990bce791b5d002a953d1f87b2ea2bc5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40129. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reasoning through video generation offers a promising path toward visual intelligence by modeling latent visual states and their dynamics. However, current video generation models often lack explicit guidance on how these states should evolve, leaving generated trajectories prone to physical and structural inconsistencies that undermine reasoning reliability. While the Video Joint-Embedding Predictive Architecture (V-JEPA) provides rich spatiotemporal priors learned through latent prediction, these general priors do not naturally adapt to the logical reasoning capabilities required for complex visual tasks. To bridge this gap, we propose VR-JEPA, a framework that aligns the V-JEPA predictor with task-specific reasoning logic through localized contrastive-state learning and uses its predicted latent trajectories to guide video generation for visual reasoning. Specifically, (i) we pair successful trajectories with generated alternatives under the same input conditions and use discrepancies in their V-JEPA representations to identify informative states and tokens for localized contrastive supervision. (ii) We further equip the V-JEPA predictor with skill-specific experts trained on anchor-task data, allowing the model to adaptively specialize its shared spatiotemporal priors across diverse cognitive domains. Together with skill-specific experts, this contrastive supervision enables VR-JEPA to predict latent trajectories that provide task-specific logical guidance for video generation. Comprehensive experiments on the large-scale VBVR-Pro-Bench dataset demonstrate that VR-JEPA achieves an $11.33\%$ relative improvement over the cutting-edge generation-based reasoning baseline, significantly mitigating physical artifacts and enhancing logical consistency.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.40129 — 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 2026-10-07T14:19:00.847Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).