aiqbench analysis
This research proposes using independently trained diffusion models as plug-and-play priors within a broader inference framework constrained by differentiable auxiliary functions. By iterating differentiation through a fixed denoising network, the method enables approximate inference for tasks like conditional generation and image segmentation while potentially offering new approaches to combinatorial optimization. The work highlights how modularizing diffusion models can adapt generative capabilities to diverse domains without retraining the core prior.
Diffusion models as plug-and-play priors
arXiv:2206.09012v4 Announce Type: replace Abstract: We consider the problem of inferring high-dimensional data $\mathbf{x}$ in a model that consists of a prior $p(\mathbf{x})$ and an auxiliary differentiable constraint $c(\mathbf{x},\mathbf{y})$ on $x$ given some additional information $\mathbf{y}$. In this paper, the prior is an independently trained denoising diffusion generative model. The auxiliary constraint is expected to have a differentiable form, but can come from diverse sources. The possibility of such inference turns diffusion models into plug-and-play modules, thereby allowing a range of potential applications in adapting models to new domains and tasks, such as conditional generation or image segmentation. The structure of diffusion models allows us to perform approximate inference by iterating differentiation through the fixed denoising network enriched with different amounts of noise at each step. Considering many noised versions of $\mathbf{x}$ in evaluation of its fitness is a novel search mechanism that may lead to new algorithms for solving combinatorial optimization problems.
Abstract:We consider the problem of inferring high-dimensional data $\mathbf{x}$ in a model that consists of a prior $p(\mathbf{x})$ and an auxiliary differentiable constraint $c(\mathbf{x},\mathbf{y})$ on $x$ given some additional information $\mathbf{y}$. In this paper, the prior is an independently trained denoising diffusion generative model. The auxiliary constraint is expected to have a differentiable form, but can come from diverse sources. The possibility of such inference turns diffusion models into plug-and-play modules, thereby allowing a range of potential applications in adapting models to new domains and tasks, such as conditional generation or image segmentation. The structure of diffusion models allows us to perform approximate inference by iterating differentiation through the fixed denoising network enriched with different amounts of noise at each step. Considering many noised versions of $\mathbf{x}$ in evaluation of its fitness is a novel search mechanism that may lead to new algorithms for solving combinatorial optimization problems.
Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2206.09012 [cs.LG]
(or arXiv:2206.09012v4 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2206.09012 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.52202/068431-1070 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 256 words
Related coverage
PSI3D: Plug-and-Play 3D Stochastic Inference with Slice-wise Latent Diffusion Prior
arXiv · Aug 6, 2026
Physics-Guided Flow Matching for CT Image Reconstruction
arXiv · Sep 1, 2026
Plug-and-Play Diffusion Meets ADMM: Dual-Variable Coupling for Robust Medical Image Reconstruction
arXiv · Jun 5, 2026
Spanning Diffusion and Flow-based Posterior Sampling via Exact Gaussian Conditioning
arXiv · Oct 12, 2026