Metraipolis › archive › dc_274887f93caa

PhyLatent: Learning Dynamics-Relevant Representations for JEPA World Models — arXiv:2608.05720

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.05720
License (as archived): CC0-1.0
Archived: 2026-10-31T10:36:09.355Z
SHA-256 of the archived text: 2ca36b4c6f09388dbffe24afc1d81325da2e8f256cd9502481653e99d94565a8
Stable URL of this page: https://metraipolis.com/documento/dc_274887f93caa

How to cite

PhyLatent: Learning Dynamics-Relevant Representations for JEPA World Models — arXiv:2608.05720 — arXiv (en), https://arxiv.org/abs/2608.05720 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_274887f93caa (Metraipolis archive dc_274887f93caa, sha256 2ca36b4c6f09388dbffe24afc1d81325da2e8f256cd9502481653e99d94565a8)

BibTeX

@misc{dc274887f93caa,
  title = {PhyLatent: Learning Dynamics-Relevant Representations for JEPA World Models — arXiv:2608.05720},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.05720},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_274887f93caa; sha256 2ca36b4c6f09388dbffe24afc1d81325da2e8f256cd9502481653e99d94565a8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05720. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We propose PhyLatent, a dynamics-relevant training objective for Joint-Embedding Predictive Architecture (JEPA) world models. Our key observation is that preventing global latent collapse does not necessarily ensure that the learned representation preserves physically meaningful state and action relationships. We identify three failure modes: sensitivity to appearance changes that leave the physical state unchanged, insufficient separation of distinct physical states, and insufficient separation of different action-conditioned futures. We refer to these as Physical Invariance Collapse, Physical Distinguishability Collapse, and Counterfactual Dynamics Collapse, respectively. PhyLatent targets these failures through three coordinated training pathways, implemented with static visual invariance, physical state grounding, future representation alignment, counterfactual branch separation, and latent denoising. On OGBench-Cube, PhyLatent reduces the three collapse rates by 43.9%, 27.9%, and 47.1%, respectively, while improving model predictive control (MPC) success by 12.0 percentage points (17.2% relative). Across four visual-control tasks, average planning success increases by 6.62 percentage points (8.3% relative). These results show that global non-collapse alone is insufficient for learning a reliable JEPA world-model state space, and that explicitly preserving dynamics-relevant structure can improve closed-loop planning.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.05720 — 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 2026-10-31T10:23:31.004Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).