标题： 生成扩散后验采样用于信息性似然 显示英文标题

标题： Generative diffusion posterior sampling for informative likelihoods

摘要： 顺序蒙特卡洛（SMC）方法最近在生成扩散模型的条件采样中展示了成功的结果。 本文提出了一种新的扩散后验SMC采样器，实现了统计效率的提升，特别是在异常值条件或高度信息量的似然函数下。 关键思想是构建一个与扩散模型相关的观测路径，并设计采样器以利用这种相关性来实现更高效的采样。 实证结果证明了效率的提升。 显示英文摘要

摘要： Sequential Monte Carlo (SMC) methods have recently shown successful results for conditional sampling of generative diffusion models. In this paper we propose a new diffusion posterior SMC sampler achieving improved statistical efficiencies, particularly under outlier conditions or highly informative likelihoods. The key idea is to construct an observation path that correlates with the diffusion model and to design the sampler to leverage this correlation for more efficient sampling. Empirical results conclude the efficiency.