提出Fresco框架，让扩散模型高效生成图像，速度提升10倍且不丢质量。
From Sketch to Fresco: Efficient Diffusion Transformer with Progressive Resolution
- 分阶段逐步升采样，避免盲目重加噪，保持各阶段结构一致
- 在FLUX上实现10倍加速，与蒸馏结合可达22倍速度提升
- 适合追求生成效率的视觉生成研究者和开发者
扩散Transformer虽生成效果出色，但因迭代采样计算开销大。近期动态分辨率采样通过降低早期步骤分辨率实现加速，但现有方法在分辨率切换时依赖启发式重加噪，破坏跨阶段一致性，迫使模型重新学习全局结构。此外，这些方法不区分收敛区域，一次性全空间上采样，导致误差累积和可见伪影。为此，我们提出Fresco，一种统一重加噪与全局结构的动态分辨率框架，采用渐进式上采样，既保留低分辨率草图的效率，又保证高分辨率精修的保真度，所有阶段对齐同一最终目标。Fresco在多个领域和模型中实现近无损加速，如FLUX上达10倍、HunyuanVideo上5倍；与蒸馏、量化、特征缓存等技术正交，结合蒸馏模型可实现22倍加速。代码见补充材料，将发布于Github。
原文摘要 · Abstract (English)
Diffusion Transformers achieve impressive generative quality but remain computationally expensive due to iterative sampling. Recently, dynamic resolution sampling has emerged as a promising acceleration technique by reducing the resolution of early sampling steps. However, existing methods rely on heuristic re-noising at every resolution transition, injecting noise that breaks cross-stage consistency and forces the model to relearn global structure. In addition, these methods indiscriminately upsample the entire latent space at once without checking which regions have actually converged, causing accumulated errors, and visible artifacts. Therefore, we propose \textbf{Fresco}, a dynamic resolution framework that unifies re-noise and global structure across stages with progressive upsampling, preserving both the efficiency of low-resolution drafting and the fidelity of high-resolution refinement, with all stages aligned toward the same final target. Fresco achieves near-lossless acceleration across diverse domains and models, including 10$\times$ speedup on FLUX, and 5$\times$ on HunyuanVideo, while remaining orthogonal to distillation, quantization and feature caching, reaching 22$\times$ speedup when combined with distilled models. Our code is in supplementary material and will be released on Github.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。