Metraipolis › archive › dc_dd8a0d638ca0

SpatialSkill: Self-Evolving Skills for Cross-View Spatial Reasoning — arXiv:2609.34124

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34124
License (as archived)
CC0-1.0
Archived
2026-10-07T23:51:52.090Z
SHA-256 of the archived text
7fb386a0f9e9c15ea87e575173cf734d9fc350cfe2d36da1fb6d0087992234a4
Stable URL of this page
https://metraipolis.com/documento/dc_dd8a0d638ca0

How to cite

SpatialSkill: Self-Evolving Skills for Cross-View Spatial Reasoning — arXiv:2609.34124 — arXiv (en), https://arxiv.org/abs/2609.34124 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dd8a0d638ca0 (Metraipolis archive dc_dd8a0d638ca0, sha256 7fb386a0f9e9c15ea87e575173cf734d9fc350cfe2d36da1fb6d0087992234a4)
BibTeX
@misc{dcdd8a0d638ca0,
  title = {SpatialSkill: Self-Evolving Skills for Cross-View Spatial Reasoning — arXiv:2609.34124},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34124},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dd8a0d638ca0; sha256 7fb386a0f9e9c15ea87e575173cf734d9fc350cfe2d36da1fb6d0087992234a4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34124. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Cross-view spatial reasoning requires a model to align different viewpoints into a coherent spatial representation, yet this ability remains challenging for vision-language models despite being natural to humans. Existing methods typically improve spatial reasoning by updating model weights, which keeps the acquired knowledge implicit and tied to a specific backbone. We propose \textit{SpatialSkill}, a weight-update-free framework that enables a frozen vision-language model to accumulate explicit natural-language reasoning skills from offline trajectories. Unlike symbolic tasks, perceptual skills cannot be reliably verified simply by executing them: a plausible spatial rule may lack visual support or require transformations that the frozen model cannot perform. SpatialSkill therefore admits candidate skills only after visual-grounding and executability checks, constrains manual evolution to prevent harmful regressions, and routes skills by spatial-reasoning category to reduce negative transfer. On CityCube, across four frozen executors, SpatialSkill yields consistent gains, and a 9B executor equipped with SpatialSkill surpasses the strongest closed-source reference in our evaluation. The skills are stored in a versioned natural-language manual, making the reasoning strategies explicit and auditable without modifying model parameters. Code at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.34124 — 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 2026-10-07T23:29:27.288Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).