Learning Quantized Adaptive Conditions for Diffusion Models
This addresses the challenge of efficient high-quality image generation for diffusion model users, representing a strong specific gain rather than incremental.
The paper tackled the problem of high curvature in ODE trajectories of diffusion models, which hinders high-quality image generation with few function evaluations, by proposing a novel approach using adaptive conditions with a quantized encoder, achieving significantly better sample quality with only 6 NFE, such as 5.14 FID on CIFAR-10.
The curvature of ODE trajectories in diffusion models hinders their ability to generate high-quality images in a few number of function evaluations (NFE). In this paper, we propose a novel and effective approach to reduce trajectory curvature by utilizing adaptive conditions. By employing a extremely light-weight quantized encoder, our method incurs only an additional 1% of training parameters, eliminates the need for extra regularization terms, yet achieves significantly better sample quality. Our approach accelerates ODE sampling while preserving the downstream task image editing capabilities of SDE techniques. Extensive experiments verify that our method can generate high quality results under extremely limited sampling costs. With only 6 NFE, we achieve 5.14 FID on CIFAR-10, 6.91 FID on FFHQ 64x64 and 3.10 FID on AFHQv2.