Metraipolis › archive › dc_567863d112e5

Accelerating Diffusion Sampling via Speculative Draft Trees — arXiv:2609.17691

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.17691
License (as archived)
CC0-1.0
Archived
2026-10-15T15:04:23.576Z
SHA-256 of the archived text
7ca62a2f0eb8b7f5b61b929b7b6a81c3cf68bb37f9684b431704e1649fa42c8e
Stable URL of this page
https://metraipolis.com/documento/dc_567863d112e5

How to cite

Accelerating Diffusion Sampling via Speculative Draft Trees — arXiv:2609.17691 — arXiv (en), https://arxiv.org/abs/2609.17691 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_567863d112e5 (Metraipolis archive dc_567863d112e5, sha256 7ca62a2f0eb8b7f5b61b929b7b6a81c3cf68bb37f9684b431704e1649fa42c8e)
BibTeX
@misc{dc567863d112e5,
  title = {Accelerating Diffusion Sampling via Speculative Draft Trees — arXiv:2609.17691},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17691},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_567863d112e5; sha256 7ca62a2f0eb8b7f5b61b929b7b6a81c3cf68bb37f9684b431704e1649fa42c8e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17691. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Speculative sampling accelerates diffusion model generation by drafting inexpensive candidate states and correcting them under a coupling that preserves the target distribution exactly, reducing the number of expensive target evaluations. Existing diffusion samplers, notably those based on reflection maximal coupling, are topologically constrained: their lookahead drafts form a chain graph, a single linear sequence, which inherently limits the acceptance rate per target evaluation. We connect speculative sampling in diffusion models to relative entropy coding (REC). This perspective shows the lookahead need not be linear and motivates our central contribution, draft trees, which enrich the candidates considered per round and lower the target function evaluations. We further adopt greedy rejection sampling, an REC algorithm, as the draft-target coupling, improving acceptance while guaranteeing exact target samples. Experiments across diverse target and draft models demonstrate up to 8.3% acceleration over the reflection coupling baseline in practical settings.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Applications (stat.AP).
Pagina del lavoro: https://arxiv.org/abs/2609.17691 — 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 2026-10-15T14:08:19.058Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).