Song and colleagues: score-based SDE and probability-flow modeling

What changed

Unified noise-conditioned generative processes through reverse-time SDEs and associated probability-flow ODEs, with predictor-corrector sampling.

Scientific scope

Unified noise-conditioned generative processes through reverse-time SDEs and associated probability-flow ODEs, with predictor-corrector sampling. Reverse-time stochastic calculus is inherited. Marginal equivalence and likelihood formulas rely on regularity and appropriate score fields; learned scores and numerical solvers add error.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Reverse-time stochastic calculus is inherited. Marginal equivalence and likelihood formulas rely on regularity and appropriate score fields; learned scores and numerical solvers add error.

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.

• Score-Based Generative Modeling through Stochastic Differential Equations

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