Schrödinger Bridge-based generative semantic communication (SBGSC) enables direct optimal distribution transport from semantics to images, cutting hallucinations and achieving 38% better FID, 49.3% better SSIM, and over 8x faster inference than prior GSC methods.
Note on the derivatives with respect to a parameter of the solutions of a system of differential equations
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
Develops safe particle flow for constrained variational inference by applying control barrier functions to probability densities with theoretical guarantees.
Under a tensor generalized detailed-balance condition, tensor-coupled flow-conservation systems on hypergraphs have a unique equilibrium with global asymptotic stability via an entropy Lyapunov function, plus sensitivity bounds and local ISS linking spectral gap to robustness.
citing papers explorer
-
Optimally Bridging Semantics and Data: Generative Semantic Communication via Schr\"odinger Bridge
Schrödinger Bridge-based generative semantic communication (SBGSC) enables direct optimal distribution transport from semantics to images, cutting hallucinations and achieving 38% better FID, 49.3% better SSIM, and over 8x faster inference than prior GSC methods.
-
Constrained Variational Inference via Safe Particle Flow
Develops safe particle flow for constrained variational inference by applying control barrier functions to probability densities with theoretical guarantees.
-
Stability and Robustness of Tensor-Coupled Flow-Conservation Dynamical Systems on Hypergraphs
Under a tensor generalized detailed-balance condition, tensor-coupled flow-conservation systems on hypergraphs have a unique equilibrium with global asymptotic stability via an entropy Lyapunov function, plus sensitivity bounds and local ISS linking spectral gap to robustness.