Consistency Regularised Gradient Flows for Inverse Problems
A Euclidean-Wasserstein-2 gradient-flow framework unifies posterior sampling and prompt optimization for low-NFE inverse problem solving using vision-language diffusion priors without backpropagating through autoencoders.
Published 2026Paris Poster Session 5 · Fri, Dec 11, 11:30 AM–1:30 PM local time · Paris Poster HallarXiv ↗OpenReview ↗
Only vote on papers you've read. Sign in with GitHub to vote.
Abstract
Vision-Language Latent Diffusion Models (LDMs) (Rombach et al., 2022) provide powerful generative priors for inverse problems. However, existing LDM-based inverse solvers typically require a large number of neural function evaluations (NFEs) and backpropagation through large pretrained components, leading to substantial computational costs and, in some cases, degraded reconstruction quality. We propose a unified Euclidean-Wasserstein-2 gradient-flow framework that jointly performs posterior sampling and prompt optimization in the latent space through a single flow that aligns the prior and posterior with the observed data. Combined with few-step latent text-to-image models, this formulation enables low-NFE inference without backpropagation through autoencoders. Experiments across several canonical imaging inverse problems show that our method achieves state-of-the-art performance with significantly reduced computational cost.