不训练即可加速扩散模型，按令牌动态选最优去噪器。
TAP: A Token-Adaptive Predictor Framework for Training-Free Diffusion Acceleration
- 用首层单次推理做探针，为每个令牌计算代理误差。
- 在不损失视觉质量前提下，速度提升显著优于固定预测器。
- 无需训练、兼容多种模型，适合追求推理效率的研究者。
扩散模型生成效果强但推理慢，因需反复进行全模型去噪。本文提出无需训练的Token-Adaptive Predictor（TAP）框架，通过一次模型首层评估作为低成本探针，计算一组候选预测器（主要基于不同阶数与时域的泰勒展开）的代理损失，再为每个令牌分配误差最小的预测器。该“探针-选择”策略利用了不同令牌的异质性时间动态特性，无需额外训练且兼容多种预测器设计。TAP开销极低，能实现大幅加速，同时保持接近无损的感知质量。在多种扩散架构和生成任务上的实验表明，相比固定全局预测器和仅缓存基线，TAP显著提升了精度-效率权衡表现。
原文摘要 · Abstract (English)
Diffusion models achieve strong generative performance but remain slow at inference due to the need for repeated full-model denoising passes. We present Token-Adaptive Predictor (TAP), a training-free, probe-driven framework that adaptively selects a predictor for each token at every sampling step. TAP uses a single full evaluation of the model's first layer as a low-cost probe to compute proxy losses for a compact family of candidate predictors (instantiated primarily with Taylor expansions of varying order and horizon), then assigns each token the predictor with the smallest proxy error. This per-token "probe-then-select" strategy exploits heterogeneous temporal dynamics, requires no additional training, and is compatible with various predictor designs. TAP incurs negligible overhead while enabling large speedups with little or no perceptual quality loss. Extensive experiments across multiple diffusion architectures and generation tasks show that TAP substantially improves the accuracy-efficiency frontier compared to fixed global predictors and caching-only baselines.
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。