Metraipolis › archive › dc_e4576cc6f122

Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models — arXiv:2608.10864

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10864
License (as archived)
CC0-1.0
Archived
2026-10-19T06:30:30.096Z
SHA-256 of the archived text
83032937cf0531fc420ca33db037c90b3af2faa661b91d8db9b6bb79b0a07757
Stable URL of this page
https://metraipolis.com/documento/dc_e4576cc6f122

How to cite

Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models — arXiv:2608.10864 — arXiv (en), https://arxiv.org/abs/2608.10864 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e4576cc6f122 (Metraipolis archive dc_e4576cc6f122, sha256 83032937cf0531fc420ca33db037c90b3af2faa661b91d8db9b6bb79b0a07757)
BibTeX
@misc{dce4576cc6f122,
  title = {Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models — arXiv:2608.10864},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10864},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e4576cc6f122; sha256 83032937cf0531fc420ca33db037c90b3af2faa661b91d8db9b6bb79b0a07757},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10864. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language models (VLMs) have achieved strong image and video understanding, yet their visual-spatial representations remain geometrically fragile, leading to failures in spatial reasoning needed for embodied AI, robotics, and autonomous driving. Prior approaches to geometry grounding either fine-tune VLMs on spatial question answering, which can perpetuate spurious visual representations, or fuse features from large geometry-grounded vision models, which substantially increases model size at inference. Knowledge distillation from geometry-grounded vision models offers an alternative, but directly matching multi-view teacher features can disrupt the pretrained alignment between visual and textual representations, degrading object- and language-semantic capabilities. We propose multi-view relational distillation (MVRD), which distills patch-wise cosine similarities across views instead of the teacher features themselves. These relations encode geometric correspondences adequate for spatial understanding, while leaving the student representation underdetermined, allowing it to remain close to its pretrained vision- language space. Across representative VLMs, MVRD improves visual-spatial reasoning, outperforming supervised fine-tuning and feature distillation while approaching feature fusion methods with considerably fewer added parameters and lower latency. We show that MVRD makes visual representations more geometric while retaining language alignment, and generalizes to 3D scene understanding tasks such as object grounding, dense captioning, and question answering.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-12 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.10864 — 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 2026-10-19T06:25:27.291Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).