Accelerating Diffusion Decoders via Multi-Scale Sampling and One-Step Distillation

TL;DR

This paper introduces a two-stage acceleration framework for diffusion decoders in image tokenization, combining multi-scale sampling and one-step distillation to significantly reduce decoding time while maintaining high image quality.

Contribution

It proposes a novel multi-scale sampling and distillation method that accelerates diffusion-based image decoders by an order of magnitude with minimal quality loss.

Findings

Achieves $ imes$10 speedup in decoding time.

Maintains high perceptual fidelity in reconstructed images.

Provides a scalable framework for efficient image tokenization.

Abstract

Image tokenization plays a central role in modern generative modeling by mapping visual inputs into compact representations that serve as an intermediate signal between pixels and generative models. Diffusion-based decoders have recently been adopted in image tokenization to reconstruct images from latent representations with high perceptual fidelity. In contrast to diffusion models used for downstream generation, these decoders are dedicated to faithful reconstruction rather than content generation. However, their iterative sampling process introduces significant latency, making them impractical for real-time or large-scale applications. In this work, we introduce a two-stage acceleration framework to address this inefficiency. First, we propose a multi-scale sampling strategy, where decoding begins at a coarse resolution and progressively refines the output by doubling the resolution at each stage, achieving a theoretical speedup of $\mathcal{O}(\log n)$ compared to standard full-resolution sampling. Second, we distill the diffusion decoder at each scale into a single-step denoising model, enabling fast and high-quality reconstructions in a single forward pass per scale. Together, these techniques yield an order-of-magnitude reduction in decoding time with little degradation in output quality. Our approach provides a practical pathway toward efficient yet expressive image tokenizers. We hope it serves as a foundation for future work in efficient visual tokenization and downstream generation.