---
title: "Accelerating posterior sampling in Bayesian inverse problems with diffusion models: an online-offline perspective"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.01300v1"
permalink: "https://feed.astralor.com/items/itm_ea6e07fa6cbe65a75d95bc32"
published: "2026-10-31T21:59:44.000Z"
score: 54
channel: "tech"
tags: ["推理效率","Accelerating","Bayesian","Score-based","Monte Carlo"]
---

# Accelerating posterior sampling in Bayesian inverse problems with diffusion models: an online-offline perspective

## At a glance

Bayesian inverse problems require both informative prior models and computationally feasible posterior inference. Score-based diffusion models provide expressive learned priors, but their use for inversion introduces a …

## Key points

- Bayesian inverse problems require both informative prior models and computationally feasible posterior inference.
- Score-based diffusion models provide expressive learned priors, but their use for inversion introduces a conditioning problem: the score of a noised posterior is generally unavailable from a prior score alone.
- This selective survey reviews the principal approaches to that problem, including likelihood guidance, spectral and projection methods, sequential Monte Carlo corrections, and amortized conditional training.

## Full text

Bayesian inverse problems require both informative prior models and computationally feasible posterior inference. Score-based diffusion models provide expressive learned priors, but their use for inversion introduces a conditioning problem: the score of a noised posterior is generally unavailable from a prior score alone. This selective survey reviews the principal approaches to that problem, including likelihood guidance, spectral and projection methods, sequential Monte Carlo corrections, and amortized conditional training. We organize methods by how they split cost between offline preparation and online inference, and compare their assumptions, advantages and limitations, with emphasis on computational speedups. Some ongoing directions and open questions are gathered at the end.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_ea6e07fa6cbe65a75d95bc32