面向扩散模型的采样感知量化
Sampling-Aware Quantization for Diffusion Models
针对扩散模型在低延迟场景下的部署，提出采样感知量化策略，通过混合阶轨迹对齐约束每步采样误差，使量化后的模型仍能保持高阶采样器的快速收敛特性，为视觉生成模型的端侧加速提供可复用方案。
原文依据 we propose a sampling-aware quantization strategy, wherein a Mixed-Order Trajectory Alignment technique is devised
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论扩散模型近期已成为视觉生成任务中的主导方法。然而，冗长的去噪链和计算密集的噪声估计网络阻碍了其在低延迟和资源受限环境中的适用性。先前的研究以解耦方式处理这些限制，要么采用先进的采样器，要么采用高效的模型量化技术。在本研究中，我们发现量化引入的噪声会破坏每个采样步的方向估计，当通过离散化数值方法求解采样方程时，进一步扭曲高阶采样器的精确方向估计，从而改变最优采样轨迹。为实现高保真度的双重加速，我们提出了一种采样感知量化策略，其中设计了一种混合阶轨迹对齐技术，以对每个采样步的误差界施加更严格的约束，促进更线性的概率流。在多个数据集上进行稀疏步快速采样的广泛实验表明，我们的方法在保持高速采样器快速收敛特性的同时，维持了优越的生成质量。代码已公开于：https://github.com/TaylorJocelyn/Sampling-aware-Quantization。
英文原文摘要 1354 字符
Diffusion models have recently emerged as the dominant approach in visual generation tasks. However, the lengthy denoising chains and the computationally intensive noise estimation networks hinder their applicability in low-latency and resource-limited environments. Previous research has endeavored to address these limitations in a decoupled manner, utilizing either advanced samplers or efficient model quantization techniques. In this study, we uncover that quantization-induced noise disrupts directional estimation at each sampling step, further distorting the precise directional estimations of higher-order samplers when solving the sampling equations through discretized numerical methods, thereby altering the optimal sampling trajectory. To attain dual acceleration with high fidelity, we propose a sampling-aware quantization strategy, wherein a Mixed-Order Trajectory Alignment technique is devised to impose a more stringent constraint on the error bounds at each sampling step, facilitating a more linear probability flow. Extensive experiments on sparse-step fast sampling across multiple datasets demonstrate that our approach preserves the rapid convergence characteristics of high-speed samplers while maintaining superior generation quality. Code is publicly available at: https://github.com/TaylorJocelyn/Sampling-aware-Quantization.