Metraipolis › archive › dc_cbbeedae08e2

Simulation-free and finite-time diffusion model — arXiv:2608.03117

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.03117
License (as archived): CC0-1.0
Archived: 2026-10-31T12:42:22.110Z
SHA-256 of the archived text: 6b2ecd450a6f0296f215eab27796c59d47328e02de6c8a40c9ed6ccb108cbefc
Stable URL of this page: https://metraipolis.com/documento/dc_cbbeedae08e2

How to cite

Simulation-free and finite-time diffusion model — arXiv:2608.03117 — arXiv (en), https://arxiv.org/abs/2608.03117 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cbbeedae08e2 (Metraipolis archive dc_cbbeedae08e2, sha256 6b2ecd450a6f0296f215eab27796c59d47328e02de6c8a40c9ed6ccb108cbefc)

BibTeX

@misc{dccbbeedae08e2,
  title = {Simulation-free and finite-time diffusion model — arXiv:2608.03117},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.03117},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cbbeedae08e2; sha256 6b2ecd450a6f0296f215eab27796c59d47328e02de6c8a40c9ed6ccb108cbefc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03117. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The performance of generative diffusion models is determined by the choice of the reference diffusion process connecting the empirical and prior distributions. Conventional approaches typically trade off simulation-free training against finite-time generation. We propose a framework for designing the reference process that achieves both simultaneously. The key idea is to prescribe tractable time-dependent conditional distributions and then construct the reference process realizing them as its marginals. This framework reveals that score matching is not fundamental to diffusion-model training but instead emerges naturally through reversal of the reference process. We further show that conditional flow matching arises as the small-noise limit of the proposed framework.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.03117 — 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 2026-10-31T12:31:30.342Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).