不训练即可加速流扩散模型生成，减少计算量同时保持高质量。
Leveraging Previous Steps: A Training-free Fast Solver for Flow Diffusion
- 利用前步结果缓存，通过泰勒展开与多项式插值复用历史信息
- 在仅10次函数求值下，CIFAR-10和LSUN-Church的FID分别降至6.75和19.49
- 无需训练，适合追求高效生成的部署场景
流扩散模型（FDMs）因生成质量高而受到关注，但当前基于常微分方程（ODE）的求解器（如欧拉法）仍需大量函数求值（NFE）以保证质量，导致生成缓慢。本文提出一种无需训练的新型流求解器，通过复用前步结果降低NFE。核心思想是建立缓存，利用泰勒展开近似ODE，并通过前步数据与多项式插值估算高阶导数，可逼近的阶数等于缓存步数。理论证明该方法具有更小近似误差与更快生成速度。在CIFAR-10、CelebA-HQ、LSUN-Bedroom、LSUN-Church、ImageNet及真实文本到图像生成任务上的实验表明其高效性：在NFE=10时，CIFAR-10的FID-30K从13.79降至6.75，LSUN-Church从46.64降至19.49。
原文摘要 · Abstract (English)
Flow diffusion models (FDMs) have recently shown potential in generation tasks due to the high generation quality. However, the current ordinary differential equation (ODE) solver for FDMs, e.g., the Euler solver, still suffers from slow generation since ODE solvers need many number function evaluations (NFE) to keep high-quality generation. In this paper, we propose a novel training-free flow-solver to reduce NFE while maintaining high-quality generation. The key insight for the flow-solver is to leverage the previous steps to reduce the NFE, where a cache is created to reuse these results from the previous steps. Specifically, the Taylor expansion is first used to approximate the ODE. To calculate the high-order derivatives of Taylor expansion, the flow-solver proposes to use the previous steps and a polynomial interpolation to approximate it, where the number of orders we could approximate equals the number of previous steps we cached. We also prove that the flow-solver has a more minor approximation error and faster generation speed. Experimental results on the CIFAR-10, CelebA-HQ, LSUN-Bedroom, LSUN-Church, ImageNet, and real text-to-image generation prove the efficiency of the flow-solver. Specifically, the flow-solver improves the FID-30K from 13.79 to 6.75, from 46.64 to 19.49 with $\text{NFE}=10$ on CIFAR-10 and LSUN-Church, respectively.
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。