Metraipolis › archive › dc_c2d6d3f09490

2D Spatial Reasoning with Adaptive Neural Cellular Automata — arXiv:2610.08518

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.08518
License (as archived)
CC0-1.0
Archived
2026-10-07T13:07:42.970Z
SHA-256 of the archived text
323c93dfb22139bb856fed96f32a968311fdf365cca843714bb90a2b7e5283ba
Stable URL of this page
https://metraipolis.com/documento/dc_c2d6d3f09490

How to cite

2D Spatial Reasoning with Adaptive Neural Cellular Automata — arXiv:2610.08518 — arXiv (en), https://arxiv.org/abs/2610.08518 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c2d6d3f09490 (Metraipolis archive dc_c2d6d3f09490, sha256 323c93dfb22139bb856fed96f32a968311fdf365cca843714bb90a2b7e5283ba)
BibTeX
@misc{dcc2d6d3f09490,
  title = {2D Spatial Reasoning with Adaptive Neural Cellular Automata — arXiv:2610.08518},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.08518},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c2d6d3f09490; sha256 323c93dfb22139bb856fed96f32a968311fdf365cca843714bb90a2b7e5283ba},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08518. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Many modern learning approaches are still struggling with spatial reasoning tasks, i.e. they lack the ability to utilize geometric information of perceived entities and their spatial relation to each other to solve problems. We introduce a novel Adaptive Neural Cellular Automata (aNCA) architecture which uses deformable convolutions to dynamically adapt the perceptive field and iteratively reason over 2D spatial relations on grid-like data structures (e.g. images). Empirical results on public benchmarks show state of the art comprehensible results with high generalization abilities for solving image based puzzles like Sudoku or finding the shortest path in a maze.

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/2610.08518 — 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 2026-10-07T12:37:32.991Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).