Metraipolis › archive › dc_1c6fe33db337
GeoLatent: Geometry-Guided Latent Structuring with Routed Optimization for 3D Reasoning — arXiv:2610.02091
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.02091
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:15:59.983Z
- SHA-256 of the archived text
2cef97c1e716b8a2fe65045f8cafddbcd7d15c6fc383534affc5e47a4eaaea41- Stable URL of this page
https://metraipolis.com/documento/dc_1c6fe33db337
How to cite
GeoLatent: Geometry-Guided Latent Structuring with Routed Optimization for 3D Reasoning — arXiv:2610.02091 — arXiv (en), https://arxiv.org/abs/2610.02091 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1c6fe33db337 (Metraipolis archive dc_1c6fe33db337, sha256 2cef97c1e716b8a2fe65045f8cafddbcd7d15c6fc383534affc5e47a4eaaea41)
BibTeX
@misc{dc1c6fe33db337,
title = {GeoLatent: Geometry-Guided Latent Structuring with Routed Optimization for 3D Reasoning — arXiv:2610.02091},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.02091},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1c6fe33db337; sha256 2cef97c1e716b8a2fe65045f8cafddbcd7d15c6fc383534affc5e47a4eaaea41},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.02091. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Despite progress in vision-language models, 3D spatial reasoning from 2D images remains challenging. Text-based methods describe intermediate geometry with discrete tokens, limiting fidelity for continuous spatial relations. Continuous latents offer richer representations, but a single latent type does not explicitly separate the cues needed across spatial tasks. Decomposed spatial latents address this by representing position, direction, and global geometry separately under geometric supervision. Yet the geometry representation can still collapse toward one dominant direction, and unrestricted attention can leave the latents underused during answer learning. We introduce GeoLatent, combining Common--Residual Geometry Alignment (CR-GEO) with routed optimization to structure the geometry states while promoting latent-mediated answer learning. CR-GEO separates shared from residual teacher geometry; routed optimization jointly trains geometry and language, temporarily directs visual answer learning through the latents, and restores full attention with geometry supervision. In controlled comparisons, CR-GEO raises geometry effective rank from 1.00 to 3.87, while blocking latent readout at the bottleneck lowers direction accuracy from 89.1% to 25.8% on 128 fixed questions. After recovery, the differentiated geometry representation and latent-mediated visual route remain available alongside direct image access. GeoLatent achieves 73.0% on SPAR-Bench and 72.1% on SPBench, outperforming previously reported methods on both. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.02091 — 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 2026-10-06T10:21:28.241Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).