Diffusion Transformers (DiTs) deliver state-of-the-art generative performance but their quadratic training cost with sequence length makes large-scale pretraining prohibitively expensive. Token dropping can reduce training cost, yet naïve strategies degrade representations, and existing methods are either parameter-heavy or fail at high drop ratios. We present SPRINT, Sparse--Dense Residual Fusion for Efficient Diffusion Transformers, a simple method that enables aggressive token dropping (up to 75%) while preserving quality. SPRINT leverages the complementary roles of shallow and deep layers: early layers process all tokens to capture local detail, deeper layers operate on a sparse subset to cut computation, and their outputs are fused through residual connections. Training follows a two-stage schedule: long masked pre-training for efficiency followed by short full-token fine-tuning to close the train--inference gap. On ImageNet-1K 256x256, SPRINT achieves 9.8x training savings with comparable FID/FDD, and at inference, its Path-Drop Guidance (PDG) nearly halves FLOPs while improving quality. These results establish SPRINT as a simple, effective, and general solution for efficient DiT training.

翻译：扩散Transformer（DiT）实现了最先进的生成性能，但其训练成本随序列长度呈二次增长，使得大规模预训练极为昂贵。令牌丢弃可降低训练成本，但简单策略会损害表征质量，现有方法要么参数量大，要么在高丢弃率下失效。本文提出SPRINT（面向高效扩散Transformer的稀疏-稠密残差融合方法），这是一种简单方法，可在保持质量的同时实现激进的令牌丢弃（高达75%）。SPRINT利用浅层与深层网络的互补作用：浅层处理所有令牌以捕捉局部细节，深层仅对稀疏子集进行计算以降低运算量，并通过残差连接融合二者输出。训练采用两阶段策略：先进行长序列掩码预训练以提升效率，再进行短序列全令牌微调以弥合训练-推理差距。在ImageNet-1K 256x256数据集上，SPRINT以可比的FID/FDD指标实现9.8倍训练成本节省；推理阶段，其路径丢弃引导（PDG）机制在提升质量的同时将FLOPs降低近半。这些结果表明SPRINT为高效DiT训练提供了一种简单、有效且通用的解决方案。