用多核并行加速扩散模型采样，不降质还提速。
CHORDS: Diffusion Sampling Accelerator with Multi-core Hierarchical ODE Solvers
- 将多核采样视为分层微分方程求解器流水线，核心间通信协同修正结果。
- 四核下最高提速2.1倍，八核达2.9倍，质量无损失。
- 无需重训练，兼容多种模型与模态，适合实时生成场景。
基于扩散的生成模型已成为高保真图像与视频生成的主流，但其推理过程计算开销巨大。现有加速方法要么需大量模型重训练，要么显著牺牲样本质量。本文提出一种通用、无需训练、模型无关的加速策略，通过多核并行实现。框架将多核扩散采样建模为一个分层常微分方程（ODE）求解器流水线，较慢但精确的求解器通过理论保证的跨核通信机制逐步修正较快求解器的结果。由此提出无需训练的扩散采样加速器CHORDS，兼容多种扩散采样器、模型架构及模态。大量实验表明，CHORDS在多个大规模图像与视频扩散模型上显著加速采样：四核下最高提速2.1倍，相比基线提升50%；八核下达2.9倍提速，且无质量损失。该进展为实时、高保真扩散生成奠定了坚实基础。
原文摘要 · Abstract (English)
Diffusion-based generative models have become dominant generators of high-fidelity images and videos but remain limited by their computationally expensive inference procedures. Existing acceleration techniques either require extensive model retraining or compromise significantly on sample quality. This paper explores a general, training-free, and model-agnostic acceleration strategy via multi-core parallelism. Our framework views multi-core diffusion sampling as an ODE solver pipeline, where slower yet accurate solvers progressively rectify faster solvers through a theoretically justified inter-core communication mechanism. This motivates our multi-core training-free diffusion sampling accelerator, CHORDS, which is compatible with various diffusion samplers, model architectures, and modalities. Through extensive experiments, CHORDS significantly accelerates sampling across diverse large-scale image and video diffusion models, yielding up to 2.1x speedup with four cores, improving by 50% over baselines, and 2.9x speedup with eight cores, all without quality degradation. This advancement enables CHORDS to establish a solid foundation for real-time, high-fidelity diffusion generation.
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。