Diffusion trajectory distillation is reframed as operator merging, yielding an optimal variance-driven merging strategy via Pareto dynamic programming in the linear Gaussian case and unavoidable approximation errors from exponential mixture growth in the nonlinear case.
Diffusion models generate images like painters: an analytical theory of outline first, details later
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
verdicts
UNVERDICTED 8representative citing papers
Analytic solution of full-batch gradient flow for linear and convolutional denoisers in diffusion models yields a universal inverse-variance spectral law for learning times of eigenmodes.
IR-guided diffusion injects intermediate text representations into early denoising steps to improve alignment for one-and-only objects, reporting up to 19.1pp VQAScore gains on OAO-AttackBench and other benchmarks.
CNS is a plug-and-play stochastic sampler for diffusion models that uses timestep- and frequency-dependent colored noise to allocate energy to unresolved bands, producing lower FID scores than standard ODE/SDE baselines on ImageNet-256.
Introduces 9 synthetic annotation tasks and benchmarks for behavioral cloning, finding hierarchical skill learning, scaling benefits, effective multi-task pretraining, and shared internal representations of task phases and mistakes.
SiLD is a score-matching framework that learns both manifold projection and intrinsic density from a single objective, with proven sample complexity depending only on intrinsic dimension.
Generative models learn rules before memorizing data, creating an innovation window whose width depends on dataset size and rule complexity, observed in both diffusion and autoregressive architectures.
DP-DMD preserves sample diversity in few-step image synthesis by applying a teacher-derived target-prediction objective to the first distillation step and standard DMD loss to the rest.
citing papers explorer
-
Toward Theoretical Insights into Diffusion Trajectory Distillation via Operator Merging
Diffusion trajectory distillation is reframed as operator merging, yielding an optimal variance-driven merging strategy via Pareto dynamic programming in the linear Gaussian case and unavoidable approximation errors from exponential mixture growth in the nonlinear case.
-
An Analytical Theory of Spectral Bias in the Learning Dynamics of Diffusion Models
Analytic solution of full-batch gradient flow for linear and convolutional denoisers in diffusion models yields a universal inverse-variance spectral law for learning times of eigenmodes.
-
Intermediate Text Representation Guided Text-to-Image Generation for Enhancing One-and-Only Alignment
IR-guided diffusion injects intermediate text representations into early denoising steps to improve alignment for one-and-only objects, reporting up to 19.1pp VQAScore gains on OAO-AttackBench and other benchmarks.
-
Colored Noise Diffusion Sampling
CNS is a plug-and-play stochastic sampler for diffusion models that uses timestep- and frequency-dependent colored noise to allocate energy to unresolved bands, producing lower FID scores than standard ODE/SDE baselines on ImageNet-256.
-
A Systematic Study of Behavioral Cloning for Scientific Data Annotation
Introduces 9 synthetic annotation tasks and benchmarks for behavioral cloning, finding hierarchical skill learning, scaling benefits, effective multi-task pretraining, and shared internal representations of task phases and mistakes.
-
Provably Learning Diffusion Models under the Manifold Hypothesis: Collapse and Refine
SiLD is a score-matching framework that learns both manifold projection and intrinsic density from a single objective, with proven sample complexity depending only on intrinsic dimension.
-
The two clocks and the innovation window: When and how generative models learn rules
Generative models learn rules before memorizing data, creating an innovation window whose width depends on dataset size and rule complexity, observed in both diffusion and autoregressive architectures.
-
Diversity-Preserved Distribution Matching Distillation for Fast Visual Synthesis
DP-DMD preserves sample diversity in few-step image synthesis by applying a teacher-derived target-prediction objective to the first distillation step and standard DMD loss to the rest.