PulseAugur
EN
LIVE 14:55:21

New theory explains flow-based solvers, proposes efficient sampling method

By PulseAugur Editorial · [5 sources] ·

Researchers have developed a new theoretical framework for understanding flow-based inverse solvers, which are used to solve imaging inverse problems. The new approach, termed posterior-transport, reveals that conditioning in these solvers is achieved through reweighting the source distribution rather than drift correction. This analysis leads to the proposal of a more efficient and principled velocity-correction solver that demonstrates competitive performance across various priors and out-of-distribution settings, while also producing diverse posterior samples with accurate uncertainty quantification. AI

IMPACT This research provides a deeper theoretical understanding of flow-based solvers and introduces a more efficient sampling method, potentially improving uncertainty quantification in inverse problem solutions.

RANK_REASON The cluster contains two arXiv papers detailing new theoretical frameworks and methods for solving inverse problems using flow-based models.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 5 sources. How we write summaries →

New theory explains flow-based solvers, proposes efficient sampling method

COVERAGE [5]

  1. arXiv cs.LG TIER_1 English(EN) · Yuanzhe Wang, Alexandre M. Tartakovsky ·

    Latent Diffusion Posterior Sampling with Surrogate Likelihood Guidance for PDE Inverse Problems

    arXiv:2606.26592v1 Announce Type: cross Abstract: We propose latent-space diffusion posterior sampling (L-DPS), an approximate Bayesian framework for high-dimensional inverse problems governed by partial differential equations (PDEs). The method addresses three challenges in PDE-…

  2. arXiv cs.LG TIER_1 English(EN) · Alexandre M. Tartakovsky ·

    Latent Diffusion Posterior Sampling with Surrogate Likelihood Guidance for PDE Inverse Problems

    We propose latent-space diffusion posterior sampling (L-DPS), an approximate Bayesian framework for high-dimensional inverse problems governed by partial differential equations (PDEs). The method addresses three challenges in PDE-constrained inversion: implicit sample-based prior…

  3. arXiv cs.CV TIER_1 English(EN) · Jian Xu, Delu Zeng, John Paisley, Qibin Zhao ·

    What Do Flow-Based Inverse Solvers Approximate? A Posterior-Transport View

    arXiv:2606.24516v1 Announce Type: new Abstract: A growing family of training-free solvers -- FlowDPS, FLOWER, PnP-Flow and their diffusion ancestors (DPS, DAPS) -- repurpose a pretrained flow-matching prior to solve imaging inverse problems by adding a measurement-guidance term t…

  4. arXiv cs.CV TIER_1 English(EN) · Qibin Zhao ·

    What Do Flow-Based Inverse Solvers Approximate? A Posterior-Transport View

    A growing family of training-free solvers -- FlowDPS, FLOWER, PnP-Flow and their diffusion ancestors (DPS, DAPS) -- repurpose a pretrained flow-matching prior to solve imaging inverse problems by adding a measurement-guidance term to the deterministic probability-flow ODE. Despit…

  5. arXiv stat.ML TIER_1 English(EN) · Yisong Yue ·

    Flow Annealing Posterior Sampling for Function-Space Regression and Inverse Problems

    Principled regression for stochastic processes is a long-standing challenge with deep connections to scientific inverse problems. We introduce Flow Annealing Posterior Sampling (FAPS), to our knowledge the first function-space posterior sampling framework that unifies stochastic-…

RELATED ENTITIES

RELATED TOPICS