通过动态补偿提升扩散模型采样速度与质量，5次评估即达顶尖性能
DC-Solver: Improving Predictor-Corrector Diffusion Sampler via Dynamic Compensation
- 引入动态补偿机制，自适应调节采样步进以解决预测-校正偏差
- 在5次函数评估下，无条件生成FID降至10.38，条件生成MSE为0.394
- 可通用增强各类扩散模型，适合追求高效高质量生成的开发者
扩散概率模型（DPM）在图像生成中表现卓越，但采样过程计算成本高，需多次函数评估。近期预测-校正采样器显著减少了函数评估次数（NFE），但因额外校正步骤存在固有偏移问题，尤其在大分类器无指导尺度（CFG）下更明显。本文提出新型快速采样器DC-Solver，通过动态补偿（DC）缓解该偏移问题。动态补偿由随采样步数自适应调整的补偿比例控制，仅需10个数据点即可通过逼近真实轨迹优化参数。进一步提出级联多项式回归（CPR），能即时预测未见配置下的补偿比例。此外，动态补偿可作为即插即用模块提升纯预测型采样器性能。大量实验表明，DC-Solver在不同分辨率（最高1024×1024）和多种DPM上均稳定超越先前方法。无条件生成在FFHQ上达10.38 FID（NFE=5），Stable-Diffusion-2.1条件下（NFE=5, CFG=7.5）MSE为0.394。代码已开源。
原文摘要 · Abstract (English)
Diffusion probabilistic models (DPMs) have shown remarkable performance in visual synthesis but are computationally expensive due to the need for multiple evaluations during the sampling. Recent predictor-corrector diffusion samplers have significantly reduced the required number of function evaluations (NFE), but inherently suffer from a misalignment issue caused by the extra corrector step, especially with a large classifier-free guidance scale (CFG). In this paper, we introduce a new fast DPM sampler called DC-Solver, which leverages dynamic compensation (DC) to mitigate the misalignment of the predictor-corrector samplers. The dynamic compensation is controlled by compensation ratios that are adaptive to the sampling steps and can be optimized on only 10 datapoints by pushing the sampling trajectory toward a ground truth trajectory. We further propose a cascade polynomial regression (CPR) which can instantly predict the compensation ratios on unseen sampling configurations. Additionally, we find that the proposed dynamic compensation can also serve as a plug-and-play module to boost the performance of predictor-only samplers. Extensive experiments on both unconditional sampling and conditional sampling demonstrate that our DC-Solver can consistently improve the sampling quality over previous methods on different DPMs with a wide range of resolutions up to 1024$\times$1024. Notably, we achieve 10.38 FID (NFE=5) on unconditional FFHQ and 0.394 MSE (NFE=5, CFG=7.5) on Stable-Diffusion-2.1. Code is available at https://github.com/wl-zhao/DC-Solver
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。