提出连续半隐式模型，加速扩散模型生成并提升性能。
Continuous Semi-Implicit Models
- 将分层半隐式模型扩展为连续框架，支持无模拟训练。
- 在图像生成任务中表现优于或相当现有加速方法。
- 适合需要高效生成与高保真度的扩散模型研究者。
半隐式分布已在变分推断和生成建模中展现巨大潜力。分层半隐式模型通过堆叠多层半隐式结构增强表达能力，可利用预训练得分网络加速扩散模型。然而，其序列化训练常导致收敛缓慢。本文提出 CoSIM，一种将分层半隐式模型拓展至连续框架的连续半隐式模型。通过引入连续转移核，CoSIM 实现无需模拟的高效训练。此外，我们证明了在精心设计的转移核下，CoSIM 可实现一致性，为生成模型在分布层面的多步蒸馏提供新思路。在图像生成上的大量实验表明，CoSIM 性能与现有扩散模型加速方法相当或更优，尤其在 FD-DINOv2 上表现突出。
原文摘要 · Abstract (English)
Semi-implicit distributions have shown great promise in variational inference and generative modeling. Hierarchical semi-implicit models, which stack multiple semi-implicit layers, enhance the expressiveness of semi-implicit distributions and can be used to accelerate diffusion models given pretrained score networks. However, their sequential training often suffers from slow convergence. In this paper, we introduce CoSIM, a continuous semi-implicit model that extends hierarchical semi-implicit models into a continuous framework. By incorporating a continuous transition kernel, CoSIM enables efficient, simulation-free training. Furthermore, we show that CoSIM achieves consistency with a carefully designed transition kernel, offering a novel approach for multistep distillation of generative models at the distributional level. Extensive experiments on image generation demonstrate that CoSIM performs on par or better than existing diffusion model acceleration methods, achieving superior performance on FD-DINOv2.
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。