Metraipolis › archive › dc_dfdb88e4f09b
Few-Step Generation via Data-Space Iteration — arXiv:2610.12102
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.12102
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T06:23:38.045Z
- SHA-256 of the archived text
1f798bb9f9cd3317e4e7b22c3f3eabc116c3474c0b8d6c9a525956a9b97059cf- Stable URL of this page
https://metraipolis.com/documento/dc_dfdb88e4f09b
How to cite
Few-Step Generation via Data-Space Iteration — arXiv:2610.12102 — arXiv (en), https://arxiv.org/abs/2610.12102 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dfdb88e4f09b (Metraipolis archive dc_dfdb88e4f09b, sha256 1f798bb9f9cd3317e4e7b22c3f3eabc116c3474c0b8d6c9a525956a9b97059cf)
BibTeX
@misc{dcdfdb88e4f09b,
title = {Few-Step Generation via Data-Space Iteration — arXiv:2610.12102},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.12102},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dfdb88e4f09b; sha256 1f798bb9f9cd3317e4e7b22c3f3eabc116c3474c0b8d6c9a525956a9b97059cf},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.12102. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Flow matching has emerged as a scalable paradigm for training high-quality generative models, but sampling from the learned probability flow requires many network evaluations. Distillation can reduce this cost to one or a few evaluations; however, one-step generation often sacrifices quality, making few-step generation the practical operating regime. Existing few-step methods perform their iterative computation along the probability flow and therefore require a fixed, manually chosen timestep discretization. This discretization is often chosen heuristically and is expensive to tune; it may also be restrictive when refinement difficulty differs across samples or spatial locations. We introduce data-space iteration, a few-step generation framework that removes flow discretization altogether. Starting from noise, a shared generator directly refines its prediction in data space, with every iteration trained to produce the best sample permitted by its capacity. Our formulation integrates with distribution matching distillation (DMD) with minimal changes, enabling a controlled comparison between iteration methods under matched training settings. On class-conditional ImageNet 256x256, data-space iteration outperforms standard discretization baselines and matches or improves upon variants selected through schedule search, without requiring schedule-specific training. These results show that data-space iteration provides a simple and effective alternative to discretized flow-space iteration for fast generation. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.12102 — 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 2026-10-09T05:50:31.705Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).