A consistency-regularized Euclidean-Wasserstein-2 gradient flow performs joint posterior sampling and prompt optimization in latent space for efficient low-NFE inverse problem solving with diffusion models.
Splitting methods for differential equations , volume =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Nesterov's acceleration is derived as the unique damping that keeps a perturbed attracting manifold tangent to the flow in a lifted phase space, with the same principle recovering both continuous and discrete versions for convex and strongly convex problems.
citing papers explorer
-
Consistency Regularised Gradient Flows for Inverse Problems
A consistency-regularized Euclidean-Wasserstein-2 gradient flow performs joint posterior sampling and prompt optimization in latent space for efficient low-NFE inverse problem solving with diffusion models.
-
On the Nesterov's acceleration: A NAIM perspective
Nesterov's acceleration is derived as the unique damping that keeps a perturbed attracting manifold tangent to the flow in a lifted phase space, with the same principle recovering both continuous and discrete versions for convex and strongly convex problems.