{
  "abstract": "Pretrained diffusion models represent image distributions through a continuum of progressively smoothed distributions. This multiscale structure organizes generation from global structure to fine detail and supports high-quality, diverse samples. We exploit the same multiscale diffusion prior for linear imaging inverse problems. Rather than using the pretrained model only as a denoiser in an outer iteration, we define a surrogate likelihood whose center is aligned with the clean-image coordinate and whose covariance accounts for residual diffusion uncertainty. This construction defines an explicit surrogate posterior path, from which we derive continuous posterior dynamics. A tunable Langevin component supports target tracking and allows the amount of posterior exploration to be adapted to the application. We prove endpoint consistency and a finite-horizon tracking bound and, in the exact-score setting, first-order weak accuracy. For computation, we derive the Posterior-Dynamics Implicit--Explicit sampler (PD-IMEX), a stable method using one score evaluation per diffusion scale and an implicit data-consistency update. Experiments on deblurring, super-resolution, and inpainting show strong reconstruction quality at 100 score evaluations, coarse-grid stability, and controllable fidelity--diversity behavior.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2608.15144",
  "author_total": 4,
  "authors": [
    {
      "name": "Zhaoqiang Liu"
    },
    {
      "name": "Tongyao Pang"
    },
    {
      "name": "Ruibing Wang"
    },
    {
      "name": "Yang Zheng"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/5rtnvmadk4dyj.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2608.15144",
  "fulltext_url": null,
  "id": "5rtnvmadk4dyj",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2608.15144",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.AI"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1707|Image and Signal Denoising Methods"
    ]
  ],
  "openalex": "W7203689795",
  "pdf_url": "https://arxiv.org/pdf/2608.15144",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/5rtnvmadk4dyj.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:18:42Z",
      "native_id": "2608.15144",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2608.15144"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors",
  "title_text": "A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors",
  "type": "preprint",
  "updated_at": "2026-10-30T12:18:42Z",
  "url": "https://research.draconitic.com/paper/5rtnvmadk4dyj",
  "venue": "arXiv",
  "versions": []
}