提出量化扩散模型的误差传播理论与补偿方法，显著提升生成质量。
Error Propagation Mechanisms and Compensation Strategies for Quantized Diffusion
- 建立扩散模型量化误差的数学传播模型，首次给出累积误差闭式解。
- 在SDXL W4A4上实现1.2分贝的PSNR提升，优于SVDQuant。
- 补偿方案仅增加不足0.5%推理时间，适合实际部署场景。
扩散模型通过建立前所未有的图像生成质量和创造力基准，彻底改变了图像合成领域。然而，其大规模部署面临计算密集的迭代去噪过程挑战。尽管训练后量化（PTQ）为加速采样提供了有效路径，但扩散模型的迭代特性导致每步量化误差逐步累积，不可避免地损害输出保真度。为此，我们构建了一个理论框架，从数学上刻画扩散模型中的误差传播机制，推导出每步量化误差传播方程，并建立了首个累积误差的闭式解。基于此理论基础，我们提出一种时间步感知的累积误差补偿方案。在多个图像数据集上的大量实验表明，该补偿策略能有效缓解误差传播，显著提升现有PTQ方法性能。具体而言，在SDXL W4A4设置下，相较SVDQuant实现1.2 PSNR提升，且额外时间开销低于0.5%。
原文摘要 · Abstract (English)
Diffusion models have transformed image synthesis by establishing unprecedented quality and creativity benchmarks. Nevertheless, their large-scale deployment faces challenges due to computationally intensive iterative denoising processes. Although post-training quantization (PTQ) provides an effective pathway for accelerating sampling, the iterative nature of diffusion models causes stepwise quantization errors to accumulate progressively during generation, inevitably compromising output fidelity. To address this challenge, we develop a theoretical framework that mathematically formulates error propagation in Diffusion Models (DMs), deriving per-step quantization error propagation equations and establishing the first closed-form solution for cumulative error. Building on this theoretical foundation, we propose a timestep-aware cumulative error compensation scheme. Extensive experiments on multiple image datasets demonstrate that our compensation strategy effectively mitigates error propagation, significantly enhancing existing PTQ methods. Specifically, it achieves a 1.2 PSNR improvement over SVDQuant on SDXL W4A4, while incurring only an additional $<$ 0.5\% time overhead.
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。