GAS: Improving Discretization of Diffusion ODEs via Generalized Adversarial Solver
This work addresses the sampling efficiency issue in diffusion models for generative AI applications, representing an incremental improvement.
The paper tackles the computationally expensive sampling problem in diffusion models by introducing the Generalized Adversarial Solver, which reduces function evaluations and improves detail fidelity, achieving superior performance over existing methods under similar resource constraints.
While diffusion models achieve state-of-the-art generation quality, they still suffer from computationally expensive sampling. Recent works address this issue with gradient-based optimization methods that distill a few-step ODE diffusion solver from the full sampling process, reducing the number of function evaluations from dozens to just a few. However, these approaches often rely on intricate training techniques and do not explicitly focus on preserving fine-grained details. In this paper, we introduce the Generalized Solver: a simple parameterization of the ODE sampler that does not require additional training tricks and improves quality over existing approaches. We further combine the original distillation loss with adversarial training, which mitigates artifacts and enhances detail fidelity. We call the resulting method the Generalized Adversarial Solver and demonstrate its superior performance compared to existing solver training methods under similar resource constraints. Code is available at https://github.com/3145tttt/GAS.