Diffusion sampling from d-dimensional distributions requires at least ~sqrt(d) adaptive score queries when score estimates have polynomial accuracy.
Gen Li, Yuting Wei, Yuxin Chen, and Yuejie Chi
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
fields
cs.LG 4verdicts
UNVERDICTED 4representative citing papers
Stable size extrapolation in local score models requires the receptive field to cover the quasi-locality range of the Gaussian-smoothed score, formalized via a size-uniform comparison theorem and validated on the new FDLF benchmark.
Stochastic interpolants unify flow-based and diffusion-based generative models by bridging target densities exactly via latent-variable processes whose drifts minimize quadratic objectives.
Full covariance matching in Gaussian DDPMs yields O(1/T^2) path KL error and is enabled by the training-free Lanczos Gaussian sampler using Jacobian-vector products.
citing papers explorer
-
Query Lower Bounds for Diffusion Sampling
Diffusion sampling from d-dimensional distributions requires at least ~sqrt(d) adaptive score queries when score estimates have polynomial accuracy.
-
When Do Local Score Models Extrapolate Across Size? A Diagnostic Theory and Benchmark
Stable size extrapolation in local score models requires the receptive field to cover the quasi-locality range of the Gaussian-smoothed score, formalized via a size-uniform comparison theorem and validated on the new FDLF benchmark.
-
Stochastic Interpolants: A Unifying Framework for Flows and Diffusions
Stochastic interpolants unify flow-based and diffusion-based generative models by bridging target densities exactly via latent-variable processes whose drifts minimize quadratic objectives.
-
The Value of Covariance Matching in Gaussian DDPMs and the Lanczos Sampler
Full covariance matching in Gaussian DDPMs yields O(1/T^2) path KL error and is enabled by the training-free Lanczos Gaussian sampler using Jacobian-vector products.