Metraipolis › archive › dc_06bbebcd8be6
Relational Abstractions for Spatial Reasoning with Diffusion Models — arXiv:2610.09780
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.09780
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T12:47:34.961Z
- SHA-256 of the archived text
86817ecf1e64a20020745005678eef44fdf5246a7d2c5eeb05479c8dc70798b4- Stable URL of this page
https://metraipolis.com/documento/dc_06bbebcd8be6
How to cite
Relational Abstractions for Spatial Reasoning with Diffusion Models — arXiv:2610.09780 — arXiv (en), https://arxiv.org/abs/2610.09780 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_06bbebcd8be6 (Metraipolis archive dc_06bbebcd8be6, sha256 86817ecf1e64a20020745005678eef44fdf5246a7d2c5eeb05479c8dc70798b4)
BibTeX
@misc{dc06bbebcd8be6,
title = {Relational Abstractions for Spatial Reasoning with Diffusion Models — arXiv:2610.09780},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.09780},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_06bbebcd8be6; sha256 86817ecf1e64a20020745005678eef44fdf5246a7d2c5eeb05479c8dc70798b4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09780. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Diffusion models excel at image synthesis, but they remain limited in their ability to reliably satisfy structured spatial reasoning constraints. In conditional data distribution modeling tasks with implicit logical structure, such as puzzles defined by visible clues paired with consistent solutions, state-of-the-art generative models tend to approximate pixel-space distributions without learning the underlying logical rules required for inference. To address this limitation, we present a novel framework for spatial reasoning with diffusion models that leverages unsupervised object discovery and abstractions of object relations. We show that the relational knowledge derived from object-centric representations enriches diffusion models with structural primitives, allowing them to effectively guide the generative representation space during both training and inference, and enabling conditional image generation that satisfies reasoning constraints. Additionally, we introduce a large-scale generative spatial reasoning benchmark with four datasets inspired by human-solvable puzzles. Our results show that relational abstractions significantly improve reasoning capabilities of diffusion models on a variety of complex reasoning tasks, while enabling robust generalization in out-of-distribution settings. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.09780 — 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 2026-10-08T12:18:15.410Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).