Metraipolis › archive › dc_dab6cbae9b35
LEF-JEPA: Denoising-Trained Latent Energy Fields for Spatial Representation Learning — arXiv:2611.08989
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.08989
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:34:42.333Z
- SHA-256 of the archived text
9427c58019288d14ec6016c39aa0a3a056be8596f5fcaa102c583cd66bee2947- Stable URL of this page
https://metraipolis.com/documento/dc_dab6cbae9b35
How to cite
LEF-JEPA: Denoising-Trained Latent Energy Fields for Spatial Representation Learning — arXiv:2611.08989 — arXiv (en), https://arxiv.org/abs/2611.08989 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dab6cbae9b35 (Metraipolis archive dc_dab6cbae9b35, sha256 9427c58019288d14ec6016c39aa0a3a056be8596f5fcaa102c583cd66bee2947)
BibTeX
@misc{dcdab6cbae9b35,
title = {LEF-JEPA: Denoising-Trained Latent Energy Fields for Spatial Representation Learning — arXiv:2611.08989},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.08989},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dab6cbae9b35; sha256 9427c58019288d14ec6016c39aa0a3a056be8596f5fcaa102c583cd66bee2947},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08989. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A joint-embedding predictive architecture (JEPA) that applies prediction supervision only at masked positions struggles to clearly capture object shapes and boundaries in its latent representations. We further observe that some existing methods improve spatial structure while also retaining rich variations in texture and color. However, preserving more low-level visual information does not necessarily make representations better suited to prediction and action. This raises a key question: how can we strengthen the spatial structure in latent representations that supports prediction and action without over-encoding these details? We introduce LEF-JEPA, which learns spatial structure through target-guided latent recovery and transfers this supervision to the visual encoder through a relative-energy constraint; the energy module is used only during pretraining and adds no inference overhead. To reduce the cost of energy optimization during pretraining, we further construct a conditional Ising Hamiltonian that allows low-energy-state search to be performed by specialized Ising solvers. On physical hardware, we validate the feasibility of achieving lower forward latency with fewer local updates. Under a unified CIFAR-10 pretraining and frozen-encoder evaluation protocol, both realizations outperform V-JEPA2.1 in image classification, foreground segmentation, and visual planning. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.08989 — 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 2026-11-09T22:30:47.312Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).