Researchers propose a novel differentiable solver search algorithm that optimizes the computational efficiency and quality of diffusion models for image generation tasks.
Diffusion models have demonstrated remarkable generation quality but at the
cost of numerous function evaluations. Recently, advanced ODE-based solvers
have been developed to mitigate the substantial computational demands of
reverse-diffusion solving under limited sampling steps. However, these solvers,
heavily inspired by Adams-like multistep methods, rely solely on t-related
Lagrange interpolation. We show that t-related Lagrange interpolation is
suboptimal for diffusion model and reveal a compact search space comprised of
time steps and solver coefficients. Building on our analysis, we propose a
novel differentiable solver search algorithm to identify more optimal solver.
Equipped with the searched solver, rectified-flow models, e.g., SiT-XL/2 and
FlowDCN-XL/2, achieve FID scores of 2.40 and 2.35, respectively, on ImageNet256
with only 10 steps. Meanwhile, DDPM model, DiT-XL/2, reaches a FID score of
2.33 with only 10 steps. Notably, our searched solver outperforms traditional
solvers by a significant margin. Moreover, our searched solver demonstrates
generality across various model architectures, resolutions, and model sizes.