Metraipolis › archive › dc_461d56666465

The role of injected noise in diffusion model generalization — arXiv:2611.03764

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03764
License (as archived)
CC0-1.0
Archived
2026-11-04T21:39:38.114Z
SHA-256 of the archived text
4a194d4177e47bdaa43e2149ffe26676cdd7803aec5147d49608454b42fa1149
Stable URL of this page
https://metraipolis.com/documento/dc_461d56666465

How to cite

The role of injected noise in diffusion model generalization — arXiv:2611.03764 — arXiv (en), https://arxiv.org/abs/2611.03764 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_461d56666465 (Metraipolis archive dc_461d56666465, sha256 4a194d4177e47bdaa43e2149ffe26676cdd7803aec5147d49608454b42fa1149)
BibTeX
@misc{dc461d56666465,
  title = {The role of injected noise in diffusion model generalization — arXiv:2611.03764},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03764},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_461d56666465; sha256 4a194d4177e47bdaa43e2149ffe26676cdd7803aec5147d49608454b42fa1149},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03764. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion models trained on distinct datasets can produce highly similar samples when prompted with the same noise input, suggesting that they learn a generalized data distribution. We investigate the role of noise injected during the sampling process, and show that it plays an important role in making or breaking this strong generalization effect. Based on this finding we investigate the robustness of sample similarity to changes in the injected noise, and find that its impact is largest in the sampling regime where broad structure already exists but the precise identity of the sample is not yet determined.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.03764 — 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 2026-11-04T21:35:49.616Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).