Gotta Go Fast When Generating Data with Score-Based Models
read the original abstract
Score-based (denoising diffusion) generative models have recently gained a lot of success in generating realistic and diverse data. These approaches define a forward diffusion process for transforming data to noise and generate data by reversing it (thereby going from noise to data). Unfortunately, current score-based models generate data very slowly due to the sheer number of score network evaluations required by numerical SDE solvers. In this work, we aim to accelerate this process by devising a more efficient SDE solver. Existing approaches rely on the Euler-Maruyama (EM) solver, which uses a fixed step size. We found that naively replacing it with other SDE solvers fares poorly - they either result in low-quality samples or become slower than EM. To get around this issue, we carefully devise an SDE solver with adaptive step sizes tailored to score-based generative models piece by piece. Our solver requires only two score function evaluations, rarely rejects samples, and leads to high-quality samples. Our approach generates data 2 to 10 times faster than EM while achieving better or equal sample quality. For high-resolution images, our method leads to significantly higher quality samples than all other methods tested. Our SDE solver has the benefit of requiring no step size tuning.
This paper has not been read by Pith yet.
Forward citations
Cited by 14 Pith papers
-
Language-Assisted Super-Resolution from Real-World Low-Resolution Patches
LA-SR redefines unpaired super-resolution in language space by projecting images into a semantically rich representation and applying vision-language model guided losses to handle real-world degradations extracted fro...
-
Lipschitz-Guided Design of Interpolation Schedules in Generative Models
Minimizing averaged squared Lipschitzness of the drift produces interpolation schedules that improve numerical accuracy and mitigate mode collapse in generative models, with closed-form optima for Gaussians and valida...
-
Diffusion and Flow Matching Models for Tabular Data: A Survey
First dedicated survey organizing diffusion and flow matching models for tabular data synthesis, imputation, anomaly detection, and related tasks, covering literature from 2015 to 2026 and highlighting open problems.
-
Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference
Latent Consistency Models enable high-fidelity text-to-image generation in 2-4 steps by directly predicting solutions to the probability flow ODE in latent space, distilled from pre-trained LDMs.
-
Diffusion Posterior Sampling for General Noisy Inverse Problems
Diffusion models solve noisy (non)linear inverse problems via approximated posterior sampling that blends diffusion steps with manifold gradients without strict consistency projection.
-
Elucidating the Design Space of Diffusion-Based Generative Models
Organizing diffusion model design choices yields SOTA FID of 1.79 on CIFAR-10 with only 35 network evaluations per image and similar gains on ImageNet-64.
-
Progressive Distillation for Fast Sampling of Diffusion Models
Progressive distillation halves sampling steps repeatedly in diffusion models, reaching 4 steps with FID 3.0 on CIFAR-10 from 8192-step samplers.
-
Compositional amortized inference for large-scale hierarchical Bayesian models
A new error-damping estimator for compositional score matching enables stable amortized inference on hierarchical Bayesian models with over 750,000 parameters using fewer than one full model simulation on large problems.
-
Improved DDIM Sampling with Moment Matching Gaussian Mixtures
Moment-matched GMM kernels in DDIM yield lower FID and higher IS than Gaussian kernels at small sampling steps on CelebA-HQ, FFHQ, ImageNet, and Stable Diffusion tasks.
-
DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models
DPM-Solver++ enables high-quality guided sampling of diffusion models in 15-20 steps via data-prediction ODE solving and multistep stabilization.
-
Language-Assisted Super-Resolution from Real-World Low-Resolution Patches
LA-SR extracts real LR patches from depth-varying regions in single images and uses vision-language models with linguistic content and quality losses for unpaired super-resolution.
-
OTCache: Optimal Transport for Geometry-Aware Caching in Diffusion Models
OTCache uses optimal transport to interpolate caching schedules between a graph-based reference and an Optuna-optimized anchor, delivering 3.66x-4.7x speedups on FLUX.1, Qwen-Image and HunyuanVideo with improved fidelity.
-
LISA: Likelihood Score Alignment for Visual-condition Controllable Generation
LISA adds a likelihood-score alignment loss to the side branch of dual-branch controllable generators, accelerating convergence and improving results across image/video tasks with negligible extra cost.
-
Mitigating the Contractivity Trap in Diffusion ODEs via Stein Stabilization
SteinDiff applies Stein-derived residual corrections at each step to stabilize large-step inference in diffusion probability flow ODEs without retraining or reference samples.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.