Metraipolis › archive › dc_d6e178ddc549
SR-JEPA: Learning Predictive Latent State in 3D Scenes — arXiv:2608.05774
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05774
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T10:36:13.641Z
- SHA-256 of the archived text
d6f2422348d434cb96dd72aff68128a01fc54a0ebec0335e701afa305aedc75f- Stable URL of this page
https://metraipolis.com/documento/dc_d6e178ddc549
How to cite
SR-JEPA: Learning Predictive Latent State in 3D Scenes — arXiv:2608.05774 — arXiv (en), https://arxiv.org/abs/2608.05774 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d6e178ddc549 (Metraipolis archive dc_d6e178ddc549, sha256 d6f2422348d434cb96dd72aff68128a01fc54a0ebec0335e701afa305aedc75f)
BibTeX
@misc{dcd6e178ddc549,
title = {SR-JEPA: Learning Predictive Latent State in 3D Scenes — arXiv:2608.05774},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05774},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d6e178ddc549; sha256 d6f2422348d434cb96dd72aff68128a01fc54a0ebec0335e701afa305aedc75f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05774. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Joint-embedding predictive architectures learn by predicting latent representations of missing observations, yet many masked JEPAs are evaluated primarily through the encoders they produce. We ask what a trained predictive pathway itself infers when an entire entity is absent from a native 3D scene. We introduce SR-JEPA, a point-native JEPA for scene-scale point clouds whose original frozen predictive pathway can be queried at a supplied location. At evaluation, every point of one object is removed before encoding and replaced by the same shape-free 32-point query at its centroid. Training uses only self-contained 3D EMA targets: no reconstruction, semantic labels, language, or lifted 2D features. On 5,953 held-out ARKitScenes objects, the imputed latent reaches 43.13% semantic-identity macro accuracy, 22.18 points above the strongest floor. Randomizing the prediction path removes 9.78 points, while substituting matched donor context removes 21.98 points. On 8,570 Sr3D support pairs, the full latent reaches 41.15 AP; identity decoded from the missing-object latent, combined with anchor identity and geometry, reaches 39.37 AP, leaving an unresolved 1.78-point residual. These results reveal a queryable, compositional 3D predictive state: the model completes context-dependent entity content, which downstream computation combines with metric geometry. 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) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.05774 — 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 2026-10-31T10:23:31.013Z 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).