Metraipolis › archive › dc_09467e5da3e1

Diffusion Models and Concept Formation — arXiv:2609.13047

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13047
License (as archived)
CC0-1.0
Archived
2026-10-16T03:02:59.614Z
SHA-256 of the archived text
f17e1f6f64203cf4731feeef0d1d40cad3b5e130a05262ffdd6aca8bf36f45ea
Stable URL of this page
https://metraipolis.com/documento/dc_09467e5da3e1

How to cite

Diffusion Models and Concept Formation — arXiv:2609.13047 — arXiv (en), https://arxiv.org/abs/2609.13047 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09467e5da3e1 (Metraipolis archive dc_09467e5da3e1, sha256 f17e1f6f64203cf4731feeef0d1d40cad3b5e130a05262ffdd6aca8bf36f45ea)
BibTeX
@misc{dc09467e5da3e1,
  title = {Diffusion Models and Concept Formation — arXiv:2609.13047},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13047},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09467e5da3e1; sha256 f17e1f6f64203cf4731feeef0d1d40cad3b5e130a05262ffdd6aca8bf36f45ea},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13047. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Humans organize knowledge into a taxonomy of concepts with nested levels of abstraction and a \emph{basic level} at which people recognize and name objects with the least cognitive effort. Cobweb is a classic cognitive account of this ability, an incremental learner that builds a probabilistic concept hierarchy by maximizing category utility. We argue that diffusion models, although designed for image synthesis, implicitly perform the same computation. The noisy marginals of a diffusion model are Gaussian smoothings of the data distribution, and the modes of these marginals form a hierarchy that corresponds to a Cobweb tree of probabilistic prototypes in four respects. Both are hierarchical density models, both are hierarchical-Bayesian models with Gaussian prototypes, both treat categorization as score-following that reduces uncertainty, and in both a basic level emerges. We locate this basic level for a diffusion model at an intermediate noise level, where recent analyses show that the reverse process commits to the class identity of a sample. The two models differ mainly in how they represent and learn the taxonomy. Cobweb learns a discrete tree incrementally, whereas a diffusion model encodes a continuous, interpolable hierarchy in a single learned score field fit to the data distribution. We test the correspondence on MNIST and Fashion-MNIST by recovering the diffusion hierarchy through mode-finding and comparing the basic levels of the two models. This reframes diffusion as a cognitive model of concept formation and offers Cobweb a continuous, scalable instantiation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-14 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.13047 — 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 2026-10-16T02:51:27.831Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).