Metraipolis › archive › dc_69a13e955028

Diffusion-free Score Matching for Exact Conditional Sampling — arXiv:2611.09758

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09758
License (as archived)
CC0-1.0
Archived
2026-11-09T22:41:41.620Z
SHA-256 of the archived text
a2a77fb8556f4bc5de1b00a01bdf116e05d142f51f141d2bc3b264abffd3ef22
Stable URL of this page
https://metraipolis.com/documento/dc_69a13e955028

How to cite

Diffusion-free Score Matching for Exact Conditional Sampling — arXiv:2611.09758 — arXiv (en), https://arxiv.org/abs/2611.09758 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_69a13e955028 (Metraipolis archive dc_69a13e955028, sha256 a2a77fb8556f4bc5de1b00a01bdf116e05d142f51f141d2bc3b264abffd3ef22)
BibTeX
@misc{dc69a13e955028,
  title = {Diffusion-free Score Matching for Exact Conditional Sampling — arXiv:2611.09758},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09758},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_69a13e955028; sha256 a2a77fb8556f4bc5de1b00a01bdf116e05d142f51f141d2bc3b264abffd3ef22},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09758. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The success of modern score-based generative models is largely driven by diffusion-based training, which learns a collection of time-indexed score fields corresponding to progressively noisier versions of the data distribution. While such models achieve impressive sample quality, leveraging them for sound probabilistic inference---particularly sampling from arbitrary conditionals of the model---remains challenging. This difficulty arises because conditioning information is only available for clean data and not for the higher noise levels required to generate exact conditional samples. In this paper, we introduce DIffusion-free SCOre matching (DISCO), a principled approach that sidesteps the need for time-dependent score fields altogether. Our method is based on an objective that, while reminiscent of diffusion-based training, estimates only a single score of the (slightly perturbed) data distribution, for which the exact conditional score is readily obtained. In experiments, DISCO is competitive with state-of-the-art diffusion models in terms of sample quality. More importantly, DISCO yields a more faithful representation of the underlying data distribution and---crucially---enables accurate sampling from arbitrary conditionals, outperforming standard heuristic samplers. This capability opens the door to principled and flexible probabilistic reasoning with score-based models.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.09758 — 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 2026-11-09T22:33:47.163Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).