Metraipolis › archive › dc_dbe77c07fb5d

A Stochastic Latent Variable Generalization of Joint-Embedding Predictive Architectures — arXiv:2611.09302

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09302
License (as archived)
CC0-1.0
Archived
2026-11-09T22:34:57.784Z
SHA-256 of the archived text
f282ed3645c3c8bed8b3e67b3d053af27f8522978b3165b972b2efa3f5248ac7
Stable URL of this page
https://metraipolis.com/documento/dc_dbe77c07fb5d

How to cite

A Stochastic Latent Variable Generalization of Joint-Embedding Predictive Architectures — arXiv:2611.09302 — arXiv (en), https://arxiv.org/abs/2611.09302 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dbe77c07fb5d (Metraipolis archive dc_dbe77c07fb5d, sha256 f282ed3645c3c8bed8b3e67b3d053af27f8522978b3165b972b2efa3f5248ac7)
BibTeX
@misc{dcdbe77c07fb5d,
  title = {A Stochastic Latent Variable Generalization of Joint-Embedding Predictive Architectures — arXiv:2611.09302},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09302},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dbe77c07fb5d; sha256 f282ed3645c3c8bed8b3e67b3d053af27f8522978b3165b972b2efa3f5248ac7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09302. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Joint-Embedding Predictive Architectures (JEPAs) predict the latent embedding of a target signal from a context signal, with an auxiliary variable z conditioning the predictor. Despite its prominent role in the original JEPA blueprint, the content of z has been carried by a fixed deterministic token in every prior instantiation, collapsing the genuinely multi-modal distribution over plausible completions to a point estimate. We present S-JEPA, which realises z as a proper stochastic latent variable for masked image prediction: a learned posterior infers z from context and target during training, while a learned context-conditioned prior approximates the distribution of plausible completions at inference, with no decoders and no reconstruction. In its deterministic limit the objective reduces to the I-JEPA loss in functional form, generalising that objective rather than strictly containing the I-JEPA model. Our central finding is that stochasticity adds capabilities, error-correlated predictive uncertainty and diverse completions, at no cost to representation quality: under matched architectures and schedules S-JEPA improves over the deterministic I-JEPA baseline and exceeds LeJEPA at matched architecture, while additionally producing spatially structured uncertainty maps and semantically diverse completions, neither of which the deterministic predictor produces.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.09302 — 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 2026-11-09T22:30:47.325Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).