Metraipolis › archive › dc_87fe615cfeae

Do Generative Priors Align with Human Naturalness Perception? — arXiv:2610.09928

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09928
License (as archived)
CC0-1.0
Archived
2026-10-08T12:47:44.625Z
SHA-256 of the archived text
fdf36815a885f9b618cef769ca52ab041ddde8010086057182a7547ffb561d1c
Stable URL of this page
https://metraipolis.com/documento/dc_87fe615cfeae

How to cite

Do Generative Priors Align with Human Naturalness Perception? — arXiv:2610.09928 — arXiv (en), https://arxiv.org/abs/2610.09928 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_87fe615cfeae (Metraipolis archive dc_87fe615cfeae, sha256 fdf36815a885f9b618cef769ca52ab041ddde8010086057182a7547ffb561d1c)
BibTeX
@misc{dc87fe615cfeae,
  title = {Do Generative Priors Align with Human Naturalness Perception? — arXiv:2610.09928},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09928},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_87fe615cfeae; sha256 fdf36815a885f9b618cef769ca52ab041ddde8010086057182a7547ffb561d1c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09928. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Visual generative models are trained to capture the probability distributions of natural images, yet whether their native priors reflect the regularities governing human perception of image naturalness remains an open question. Here, we probe these priors through native prediction errors across 25 open image and video generators. Because raw single-image losses are dominated by scene content and visual complexity, we evaluate directional loss differences using content-preserving, paired relational interventions that selectively disrupt facial configurations or physical illumination consistency while limiting changes in low-level image statistics. Across both domains, these loss differences reproduce human-like selective sensitivities and tolerances, capturing the classic Thatcher effect on faces and shape-dependent responses to illumination inconsistencies. Notably, these loss differences reliably track continuous gradations of human naturalness judgments across individual stimulus pairs (peaking at $r = .84$ on faces and $.64$ on physical scenes) and retain unique human-aligned signals even after controlling for feature distances from frozen vision encoders and standard image quality metrics. We also find that while overall sensitivity to these violations broadly covaries with human alignment across models, the two systematically decouple along denoising schedules, with alignment peaking earlier than sensitivity, revealing that human-like naturalness judgments dissociate from generic violation detection. Together, these findings demonstrate that learning visual distributions yields generative loss landscapes that capture distinct aspects of human naturalness perception.

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) ; Neurons and Cognition (q-bio.NC).
Pagina del lavoro: https://arxiv.org/abs/2610.09928 — 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 2026-10-08T12:18:15.435Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).