Metraipolis › archive › dc_968db3d1828a

Optimize Your Sampling: Tuned Diffusion Sampling with Bayesian Optimization — arXiv:2608.18040

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.18040
License (as archived)
CC0-1.0
Archived
2026-10-18T14:48:48.715Z
SHA-256 of the archived text
f83da2f19d03c74f4c00a1682d9df0b3b0fd36ef53a9ace1cd25f99f883d8b05
Stable URL of this page
https://metraipolis.com/documento/dc_968db3d1828a

How to cite

Optimize Your Sampling: Tuned Diffusion Sampling with Bayesian Optimization — arXiv:2608.18040 — arXiv (en), https://arxiv.org/abs/2608.18040 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_968db3d1828a (Metraipolis archive dc_968db3d1828a, sha256 f83da2f19d03c74f4c00a1682d9df0b3b0fd36ef53a9ace1cd25f99f883d8b05)
BibTeX
@misc{dc968db3d1828a,
  title = {Optimize Your Sampling: Tuned Diffusion Sampling with Bayesian Optimization — arXiv:2608.18040},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.18040},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_968db3d1828a; sha256 f83da2f19d03c74f4c00a1682d9df0b3b0fd36ef53a9ace1cd25f99f883d8b05},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18040. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sampling from a diffusion model typically requires many forward passes through a large neural network, making generation computationally expensive. While much work has focused on efficient solvers and samplers, comparatively little attention has been paid to selecting the sampling timesteps themselves. A recent line of work optimizes theoretically derived surrogates for sample quality rather than the quality metric itself. We propose Optimizing Your Sampling (OYS), which instead treats timestep selection as a black-box optimization problem, optimizing the target metric directly with Bayesian optimization. OYS outperforms both the default schedules and those of Align Your Steps on text-to-image generation, and improves over the default schedules on inpainting and other image tasks, in both quantitative and human evaluations. OYS requires no additional training, is applicable even to distilled models, and improves both simple and sophisticated samplers such as Euler and DPM-Solver++. A 5-step OYS schedule retains 89%-94% of the quality of a 50-step schedule while reducing inference cost by 10x.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-19 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.18040 — 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 2026-10-18T14:47:59.426Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-19?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).