2 papers
cs.LG2026
A Stabilized Path-Space Approach to Diffusion-Based Posterior Sampling
Evan Scope Crafts, Umberto Villa, Saviz Mowlavi +3
Diffusion models provide expressive data-driven priors for Bayesian inverse problems, but many diffusion posterior samplers rely on heuristic guidance approximations that can fail…
math.OC2025
Benchmarking Diffusion Annealing-Based Bayesian Inverse Problem Solvers
Evan Scope Crafts, Umberto Villa
In recent years, the ascendance of diffusion modeling as a state-of-the-art generative modeling approach has spurred significant interest in their use as priors in Bayesian inverse…