Metraipolis › archive › dc_de368be7ff4a

FLINT: Coupling Proximal Initialization and Bounded Stochasticity for Flow-Matching Inverse Problems — arXiv:2611.05945

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.05945
License (as archived)
CC0-1.0
Archived
2026-11-05T18:29:49.497Z
SHA-256 of the archived text
f868abe85f4db347e7138dd70a6dd88b81dab3b0120a4d6bbd05494f6c3b5638
Stable URL of this page
https://metraipolis.com/documento/dc_de368be7ff4a

How to cite

FLINT: Coupling Proximal Initialization and Bounded Stochasticity for Flow-Matching Inverse Problems — arXiv:2611.05945 — arXiv (en), https://arxiv.org/abs/2611.05945 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_de368be7ff4a (Metraipolis archive dc_de368be7ff4a, sha256 f868abe85f4db347e7138dd70a6dd88b81dab3b0120a4d6bbd05494f6c3b5638)
BibTeX
@misc{dcde368be7ff4a,
  title = {FLINT: Coupling Proximal Initialization and Bounded Stochasticity for Flow-Matching Inverse Problems — arXiv:2611.05945},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.05945},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_de368be7ff4a; sha256 f868abe85f4db347e7138dd70a6dd88b81dab3b0120a4d6bbd05494f6c3b5638},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05945. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Flow matching models have emerged as a dominant paradigm for generative posterior sampling in imaging inverse problems. Existing solvers follow two distinct lines of work: trajectory-space methods, which refine sampling dynamics from random initial noise, and source-space methods, which optimize the initial noise itself. While most prior work treats these as independent design choices, we show that they are inherently coupled. We derive a unified stability bound that links initialization proximity and cumulative sampling stochasticity through a multiplicative envelope, complemented by two empirical observations. First, we identify a sufficiency regime in which a moderately optimized proximal source initialization is enough to attain the performance plateau, rendering exhaustive source-space optimization unnecessary. Second, this initial advantage is fragile: it is eroded by excessive stochasticity accumulation during guided sampling, necessitating a mechanism that bounds the cumulative noise injection. These findings motivate FLINT, a two-stage framework that operationalizes this coupling. FLINT establishes a high-fidelity starting point via a manifold-constrained proximal objective in only a few iterations, then performs guided sampling under a bounded stochasticity schedule that helps prevent erasure of the initialization advantage. Across diverse inverse problems, FLINT delivers state of the art perceptual quality and competitive distortion metrics while maintaining superior computational efficiency.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.05945 — 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 2026-11-05T18:19:49.044Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).