Composing diffusion priors with explicit physical context via generative Gibbs sampling

TL;DR

GG-PA is a training-free framework that combines pretrained diffusion priors with explicit physical context through a Gibbs sampling approach, enabling accurate scientific sampling without retraining.

Contribution

Introduces GG-PA, a novel Gibbs sampling method for integrating learned priors with physical context, proven to be asymptotically exact and effective in complex physical systems.

Findings

GG-PA recovers distribution shifts induced by physical context.

It captures emergent collective behavior in interacting systems.

The method accelerates mixing via replica exchange over diffusion time.

Abstract

Pretrained diffusion models provide powerful learned priors, but in scientific sampling the target distribution often depends on physical context that is not fully represented by one generative model. We introduce Generative Gibbs for Physics-Aware Sampling (GG-PA), a training-free framework that formulates the composition of learned partial priors and explicit physical context as inference over a joint target distribution in an augmented state space. We derive a Gibbs sampler for this joint target, show that it is asymptotically exact as the diffusion time approaches zero, and prove that in settings with quadratic interactions it remains exact at finite diffusion times. We further introduce replica exchange over diffusion time to accelerate mixing. Experiments on a double-well system, a $\phi^4$ lattice model, and atomistic peptide systems show that GG-PA recovers context-induced distribution shifts and emergent collective behavior in interacting systems using partial priors without retraining. These results demonstrate GG-PA as a practical approach for combining pretrained generative priors with explicit physical context.