Metraipolis › archive › dc_157df91d2b3d
Lie Routing Vision Language Models for View-Robust 3D Understanding — arXiv:2611.02306
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02306
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:40:42.578Z
- SHA-256 of the archived text
d42ad0a4ba8d8a44a27d794922d64ffeac63a467b9bc3d73db7900f5844416b3- Stable URL of this page
https://metraipolis.com/documento/dc_157df91d2b3d
How to cite
Lie Routing Vision Language Models for View-Robust 3D Understanding — arXiv:2611.02306 — arXiv (en), https://arxiv.org/abs/2611.02306 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_157df91d2b3d (Metraipolis archive dc_157df91d2b3d, sha256 d42ad0a4ba8d8a44a27d794922d64ffeac63a467b9bc3d73db7900f5844416b3)
BibTeX
@misc{dc157df91d2b3d,
title = {Lie Routing Vision Language Models for View-Robust 3D Understanding — arXiv:2611.02306},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02306},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_157df91d2b3d; sha256 d42ad0a4ba8d8a44a27d794922d64ffeac63a467b9bc3d73db7900f5844416b3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02306. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal Large Language Models (MLLMs) have made substantial progress in 3D scene understanding from multi-view RGB observations. Yet a fixed object and query can elicit inconsistent target-related responses across camera viewpoints, a phenomenon we characterize as viewpoint-induced semantic drift. We introduce LieVLM, a Lie routing framework for view-robust 3D understanding. Motivated by the rotational component of viewpoint change, Lie routing adapts visual representations by applying local $\mathrm{SO}(3)$ transformations to projected features within learned three-dimensional subspaces. Geometry tokens extracted by VGGT predict local rotation parameters and channel-wise routing coefficients. A closed-form Padé--Cayley map transforms projected visual features, whose identity-relative corrections are projected back and gated into the original visual and geometry features. This design retains the pretrained semantic and direct spatial context while learning view-dependent feature adjustments. Experiments on 3D captioning, grounding, detection, and spatial reasoning show consistent improvements over additive fusion across training scales and backbones. Controlled ablations examine the roles of geometric conditioning and transformation structure. Viewpoint analyses further reveal systematic variation in the predicted routing states alongside more localized target responses and stronger task performance across viewpoint changes.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.02306 — 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 2026-11-03T18:12:54.764Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).