PNDM: pseudo-numerical diffusion sampling with nonlinear transfer

What changed

Separated numerical updates into estimated-gradient and transfer parts, combining multistep noise estimates with a DDIM-related nonlinear transfer to accelerate sampling from pretrained diffusion models.

Scientific scope

Separated numerical updates into estimated-gradient and transfer parts, combining multistep noise estimates with a DDIM-related nonlinear transfer to accelerate sampling from pretrained diffusion models. Inherits DDIM and classical multistep/Runge–Kutta ideas. The high-density-manifold interpretation does not certify that finite learned-model trajectories stay on an exact data manifold. Speed and error orders require their stated schedule/smoothness conditions; no universal fixed-step quality guarantee.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits DDIM and classical multistep/Runge–Kutta ideas. The high-density-manifold interpretation does not certify that finite learned-model trajectories stay on an exact data manifold. Speed and error orders require their stated schedule/smoothness conditions; no universal fixed-step quality guarantee.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Pseudo Numerical Methods for Diffusion Models on Manifolds

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction