提出误差优化缓存机制，显著提升扩散Transformer生成速度与质量。
Accelerating Diffusion Transformer via Error-Optimized Cache
- 通过提取缓存差异并动态判断优化需求，精准控制缓存过程。
- 在ImageNet上缓存率75%时FID降低28.8%，生成质量大幅提高。
- 适合追求高效高质量图像生成的研究者和开发者使用。
扩散Transformer（DiT）是内容生成的关键方法，但采样耗时较长。现有缓存方法通过复用前一时刻的DiT特征并跳过后续计算来加速，但常忽略缓存引入的误差，导致缓存强度增加时生成质量急剧下降。为此，本文提出误差优化缓存（EOC）方法，包含三项改进：(1) 先验知识提取，处理缓存差异；(2) 缓存优化判断机制，确定是否需优化特定缓存步骤；(3) 缓存优化策略，降低缓存误差。实验表明，该方法显著减少缓存引起的误差累积，尤其在高缓存强度下表现突出。在ImageNet数据集上，当基于规则的FORA模型缓存率为75%、50%、25%，基于训练的Learning-to-cache模型缓存率为22%时，生成图像的FID值分别从30.454降至21.690（↓28.8%）、6.857降至5.821（↓15.1%）、3.870降至3.692（↓4.6%）、3.539降至3.451（↓2.5%），且计算开销基本不变。代码已开源。
原文摘要 · Abstract (English)
Diffusion Transformer (DiT) is a crucial method for content generation. However, it needs a lot of time to sample. Many studies have attempted to use caching to reduce the time consumption of sampling. Existing caching methods accelerate generation by reusing DiT features from the previous time step and skipping calculations in the next, but they tend to locate and cache low-error modules without focusing on reducing caching-induced errors, resulting in a sharp decline in generated content quality when increasing caching intensity. To solve this problem, we propose the \textbf{E}rror-\textbf{O}ptimized \textbf{C}ache (\textbf{EOC}). This method introduces three key improvements: \textbf{(1)} Prior knowledge extraction: Extract and process the caching differences; \textbf{(2)} A judgment method for cache optimization: Determine whether certain caching steps need to be optimized; \textbf{(3)} Cache optimization: reduce caching errors. Experiments show that this algorithm significantly reduces the error accumulation caused by caching, especially excessive caching. On the ImageNet dataset, without substantially increasing the computational load, this method improves the FID of the generated images when the rule-based model FORA has a caching level of \textbf{75}\%, \textbf{50}\%, and \textbf{25}\%, and the training-based model Learning-to-cache has a caching level of \textbf{22}\%. Specifically, the FID values change from 30.454 to 21.690 (\textbf{28.8}\%), from 6.857 to 5.821 (\textbf{15.1}\%), from 3.870 to 3.692 (\textbf{4.6}\%), and from 3.539 to 3.451 (\textbf{2.5}\%) respectively. Code is available at https://github.com/qiujx0520/EOC_MM2025.git.
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。