Metraipolis › archive › dc_ff9be408ae7e

Geometry-Aware Diffusion Guidance via Curvature-Adaptive Tubular Correction — arXiv:2609.21251

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21251
License (as archived)
CC0-1.0
Archived
2026-10-10T02:23:02.075Z
SHA-256 of the archived text
5a650519f4a81f72ac3b0c4eadb03f37a994eaf648af2660bb0445bc53e2c953
Stable URL of this page
https://metraipolis.com/documento/dc_ff9be408ae7e

How to cite

Geometry-Aware Diffusion Guidance via Curvature-Adaptive Tubular Correction — arXiv:2609.21251 — arXiv (en), https://arxiv.org/abs/2609.21251 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff9be408ae7e (Metraipolis archive dc_ff9be408ae7e, sha256 5a650519f4a81f72ac3b0c4eadb03f37a994eaf648af2660bb0445bc53e2c953)
BibTeX
@misc{dcff9be408ae7e,
  title = {Geometry-Aware Diffusion Guidance via Curvature-Adaptive Tubular Correction — arXiv:2609.21251},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21251},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff9be408ae7e; sha256 5a650519f4a81f72ac3b0c4eadb03f37a994eaf648af2660bb0445bc53e2c953},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21251. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Gradient-guided diffusion samplers provide flexible priors for inverse problems and conditional generation, but strong guidance can move the sampling trajectory into regions where the learned score is poorly supported. Existing tangent-projection strategies limit first-order departure from an iso-density surface, yet discard potentially useful normal motion and overlook the second-order departure induced by tangent motion on a curved surface. We introduce curvature-adaptive tubular correction (CAT), a training-free plugin that regulates both effects within a shared, noise-dependent geometric budget. CAT decomposes the guidance gradient into normal and tangent components, charges normal displacement at first order and tangent displacement according to directional curvature, and obtains their jointly optimal magnitudes from a one-dimensional dual equation. Armijo backtracking calibrates the resulting finite step against the actual guidance objective, while matrix-free directional derivatives avoid constructing the full score Jacobian. We establish local guarantees for the tubular approximation, uniqueness of the correction, and sufficient objective decrease. Across seven inverse problems on FFHQ and ImageNet, CAT improves the evaluated pixel- and latent-space host samplers, with particularly consistent gains in perceptual metrics. It also improves black hole reconstruction on InverseBench and yields the lowest FID among the compared methods at every tested classifier-free guidance scale, while maintaining stable saturation and contrast. These results support curvature-aware tubular control as a reusable mechanism for stabilizing diffusion guidance.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-21 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.21251 — 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 2026-10-10T02:07:36.022Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).