Metraipolis › archive › dc_179c211ca52f

Spatially Adaptive Noise Injection — arXiv:2609.18466

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18466
License (as archived)
CC0-1.0
Archived
2026-10-15T15:04:44.738Z
SHA-256 of the archived text
b1167da658fa1ef605db80ef8158c934baf33540573ce42f75cafa737361ddbc
Stable URL of this page
https://metraipolis.com/documento/dc_179c211ca52f

How to cite

Spatially Adaptive Noise Injection — arXiv:2609.18466 — arXiv (en), https://arxiv.org/abs/2609.18466 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_179c211ca52f (Metraipolis archive dc_179c211ca52f, sha256 b1167da658fa1ef605db80ef8158c934baf33540573ce42f75cafa737361ddbc)
BibTeX
@misc{dc179c211ca52f,
  title = {Spatially Adaptive Noise Injection — arXiv:2609.18466},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18466},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_179c211ca52f; sha256 b1167da658fa1ef605db80ef8158c934baf33540573ce42f75cafa737361ddbc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18466. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion samplers reverse a learned noising process using either stochastic (DDPM) or deterministic (DDIM) updates, which represent endpoints of a single family controlled by a scalar noise-injection variance that is applied identically at every spatial location. This uniform approach neglects the geometry of natural images: high-curvature regions such as edges and textures, where the denoiser is uncertain, benefit from stochastic correction, whereas smooth regions, where the score is precise, are degraded by injected noise. This work investigates whether each pixel requires stochastic correction at a given timestep and introduces Spatially Adaptive Noise Injection (SANI), a novel sampling framework that dynamically adjusts noise application on a per-pixel basis. SANI integrates a probabilistic gating mechanism with a derived spatially adaptive variance, ensuring that noise is injected precisely where needed to refine complex features while preserving well-formed structures. Experimental results and decoupling ablations demonstrate that SANI consistently improves Fréchet Inception Distance (FID) over the vanilla DDPM and DDIM endpoint samplers across diverse sampling timesteps, while remaining competitive with variance-learning baselines, highlighting the importance of spatial adaptivity in diffusion sampling.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2609.18466 — 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 2026-10-15T14:08:19.125Z 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).