Metraipolis › archive › dc_0285887bc18d
Rethinking Pixel Mean Flows via Interval Denoiser — arXiv:2608.04818
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.04818
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T11:48:10.835Z
- SHA-256 of the archived text
b72bdf05c97726844c0559ea84cb538b770d5faa7ad3ec3250b6f64443fbf0ba- Stable URL of this page
https://metraipolis.com/documento/dc_0285887bc18d
How to cite
Rethinking Pixel Mean Flows via Interval Denoiser — arXiv:2608.04818 — arXiv (en), https://arxiv.org/abs/2608.04818 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0285887bc18d (Metraipolis archive dc_0285887bc18d, sha256 b72bdf05c97726844c0559ea84cb538b770d5faa7ad3ec3250b6f64443fbf0ba)
BibTeX
@misc{dc0285887bc18d,
title = {Rethinking Pixel Mean Flows via Interval Denoiser — arXiv:2608.04818},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.04818},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0285887bc18d; sha256 b72bdf05c97726844c0559ea84cb538b770d5faa7ad3ec3250b6f64443fbf0ba},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04818. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Modern diffusion and flow-based models are increasingly moving toward few-step, latent-free generation to bypass the computational overhead of multi-step sampling and the reconstruction bottlenecks of external autoencoders. We propose the Interval Denoiser, a theoretically rigorous framework for latent-free generation. Derived directly from the flow matching ODE, it establishes an exact analytical mapping for intermediate trajectory states. Unlike prior formulations, our prediction is shown to reside on a low-dimensional manifold across any time interval, making the regression tractable for a network operating directly on pixels. Furthermore, by avoiding empirical algebraic substitutions, our formulation correctly isolates the pure time derivative to prevent biased gradient evaluations and ensure exact first-order optimization. By analyzing this objective, we equip our framework with residual clipping and a time-sampling curriculum, enabling effective long-interval training and improving few-step performance. Trained from scratch on ImageNet 256x256, our model achieves an FID of 4.55 in one step (1-NFE) and 3.98 in two steps (2-NFE) without perceptual losses. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-06 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.04818 — 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 2026-10-31T11:37:40.719Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).