AI RESEARCH
Diffusion-Based Posterior Sampling: A Feynman-Kac Analysis of Bias and Stability
arXiv CS.LG
•
ArXi:2605.06538v1 Announce Type: new Diffusion-based posterior samplers use pretrained diffusion priors to sample from measurement- or reward-conditioned posteriors, and are widely used for inverse problems. Yet their theoretical behavior remains poorly understood: even with exact prior scores, their outputs are biased, and in low-temperature regimes their discretizations can become unstable. We characterize this bias by We apply this framework to DPS and STSL, a related sampler.