Adaptive Stochastic Coefficients for Accelerating Diffusion Sampling
This work addresses the problem of accelerating diffusion sampling for generative modeling, offering an incremental improvement by integrating a learnable coefficient into existing solvers.
The paper tackles the trade-off between computational speed and sample quality in diffusion-based generative processes by introducing AdaSDE, a novel single-step SDE solver that dynamically regulates error correction strength, achieving state-of-the-art FID scores of 4.18 on CIFAR-10, 8.05 on FFHQ, and 6.96 on LSUN Bedroom at 5 NFE.
Diffusion-based generative processes, formulated as differential equation solving, frequently balance computational speed with sample quality. Our theoretical investigation of ODE- and SDE-based solvers reveals complementary weaknesses: ODE solvers accumulate irreducible gradient error along deterministic trajectories, while SDE methods suffer from amplified discretization errors when the step budget is limited. Building upon this insight, we introduce AdaSDE, a novel single-step SDE solver that aims to unify the efficiency of ODEs with the error resilience of SDEs. Specifically, we introduce a single per-step learnable coefficient, estimated via lightweight distillation, which dynamically regulates the error correction strength to accelerate diffusion sampling. Notably, our framework can be integrated with existing solvers to enhance their capabilities. Extensive experiments demonstrate state-of-the-art performance: at 5 NFE, AdaSDE achieves FID scores of 4.18 on CIFAR-10, 8.05 on FFHQ and 6.96 on LSUN Bedroom. Codes are available in https://github.com/WLU-wry02/AdaSDE.