PyramidalWan：让预训练视频模型金字塔化以实现高效推理
PyramidalWan: On Making Pretrained Video Model Pyramidal for Efficient Inference
把已预训练的视频扩散模型通过低成本微调改造成金字塔式多分辨率去噪结构，可直接复用于任何多步扩散采样器以降低推理算力；其金字塔内步数蒸馏策略对比也为少步视频生成提供可借鉴的加速路线。
原文依据 we present a pipeline that converts a pretrained diffusion model into a pyramidal one through low-cost finetuning
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期提出的金字塔模型将传统的正向与反向扩散过程分解为在多种分辨率下运行的多个阶段。这类模型在较低分辨率下处理噪声水平较高的输入，而在较高分辨率下处理噪声较少的输入。这种分层方式显著降低了多步去噪模型推理的计算开销。然而，现有的开源金字塔视频模型均从零开始训练，在视觉合理性方面往往不及最优系统。在本工作中，我们提出一套流程，通过低成本微调将预训练扩散模型转换为金字塔模型，在实现该转换的同时不损失输出视频的质量。此外，我们研究并比较了金字塔模型内多种步数蒸馏策略，以进一步提升推理效率。
英文原文摘要 952 字符
Recently proposed pyramidal models decompose the conventional forward and backward diffusion processes into multiple stages operating at varying resolutions. These models handle inputs with higher noise levels at lower resolutions, while less noisy inputs are processed at higher resolutions. This hierarchical approach significantly reduces the computational cost of inference in multi-step denoising models. However, existing open-source pyramidal video models have been trained from scratch and tend to underperform compared to state-of-the-art systems in terms of visual plausibility.In this work, we present a pipeline that converts a pretrained diffusion model into a pyramidal one through low-cost finetuning, achieving this transformation without degradation in quality of output videos.Furthermore, we investigate and compare various strategies for step distillation within pyramidal models, aiming to further enhance the inference efficiency.