用可学习的分解缓存加速扩散模型，5-7倍提速仍保持近无损质量
LinCa: Accelerating Diffusion Models via Learnable Decomposed Feature Caching
- 通过可逆网络分解特征，按不同连续性分组预测
- 仅增0.2%参数，在5-7倍加速下保持高质量生成
- 适合需要高速图像/视频生成的部署场景
扩散模型在图像与视频生成中取得显著成功，但迭代采样带来的高计算开销仍是实际部署的关键瓶颈。特征缓存作为一种有前景的加速范式，通过复用或预测时间步间的中间特征来降低计算量。然而，现有免训练方法采用统一预测策略，无法适应特征动态的异质性，在高加速比下导致显著质量下降。本文提出LinCa，一种基于可学习可逆网络的特征缓存框架。LinCa通过轻量级可逆网络将缓存特征分解为具有不同连续性特性的子成分，并为每类成分匹配差异化的预测顺序。严格的可逆性保证了可无损重构回原始特征空间，形成统一的‘分解-预测-重构’流程。通过为不同模型和时间步段分别训练预测器，LinCa能自适应异质特征动态。在FLUX、Qwen-Image和HunyuanVideo上的实验表明，LinCa仅增加不足0.2%参数量，显著优于现有方法，在5-7倍加速下仍保持近无损质量。
原文摘要 · Abstract (English)
Diffusion models have achieved remarkable success in image and video generation, yet the high computational cost of iterative sampling remains a critical bottleneck for practical deployment. Feature caching has emerged as a promising acceleration paradigm by reusing or predicting intermediate features across timesteps. However, existing training-free methods apply uniform prediction strategies that cannot adapt to the heterogeneous feature dynamics, causing significant quality degradation under high acceleration ratios. We propose LinCa, a feature caching framework based on learnable invertible networks. LinCa decomposes cached features into sub-components with distinct continuity properties via a lightweight invertible network and applies differentiated prediction orders matched to each component. The strict invertibility guarantees lossless reconstruction back to the original feature space, forming a unified Decompose-Predict-Reconstruct pipeline. By training separate predictors for different models and timestep segments, LinCa adapts to heterogeneous feature dynamics. Experiments on FLUX, Qwen-Image, and HunyuanVideo demonstrate that LinCa, with less than 0.2% additional parameters, significantly outperforms existing methods and maintains near-lossless quality at 5-7x speedup. Code: https://github.com/QHR69/LinCa
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。