Metraipolis › archive › dc_fe53cd92028b

SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem — arXiv:2609.07064

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07064
License (as archived)
CC0-1.0
Archived
2026-10-16T15:11:37.329Z
SHA-256 of the archived text
4c26906b5587c16a02bf5cc70459dc60d1a75009dd5b7efebc0c77a6eb8cf20b
Stable URL of this page
https://metraipolis.com/documento/dc_fe53cd92028b

How to cite

SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem — arXiv:2609.07064 — arXiv (en), https://arxiv.org/abs/2609.07064 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fe53cd92028b (Metraipolis archive dc_fe53cd92028b, sha256 4c26906b5587c16a02bf5cc70459dc60d1a75009dd5b7efebc0c77a6eb8cf20b)
BibTeX
@misc{dcfe53cd92028b,
  title = {SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem — arXiv:2609.07064},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07064},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fe53cd92028b; sha256 4c26906b5587c16a02bf5cc70459dc60d1a75009dd5b7efebc0c77a6eb8cf20b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07064. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Vision-Language Models (LVLMs) have achieved strong performance on diverse visual tasks, yet their ability to reconstruct and reason about the 3D structure of the scene depicted in 2D images -- referred to as spatial intelligence -- remains limited. Existing approaches attempt to address this gap by using real-scene spatial question answering datasets that require dense geometric annotations. However, constructing such labels is costly, time-consuming, and often noisy due to reliance on external perception modules. In this work, we propose a novel paradigm inspired by human cognitive development: learning foundational spatial skills through structured block-manipulation tasks. We introduce SpatialBlock-15k, a synthetic dataset of 15,000 block-stacking problems covering 3D-to-2D projection, viewpoint transformation, and structural combination. The dataset further incorporates controlled color modulation as visual cues to encourage anchor-based reasoning in visually complex conditions. Experiments demonstrate that LVLMs trained on our dataset through either direct answering or reasoning-based prediction significantly outperform baselines and generalize to real-world spatial tasks, despite the dataset's synthetic and compact nature. Code and data are available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.07064 — 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 2026-10-16T14:49:57.174Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).