While achieving exceptional generative quality, modern diffusion, flow, and other matching models suffer from slow inference, as they require many steps of iterative generation. Recent distillation methods address this by training efficient one-step generators under the guidance of a pre-trained teacher model. However, these methods are often constrained to only one specific framework, e.g., only to diffusion or only to flow models. Furthermore, these methods are naturally data-free, and to benefit from the usage of real data, it is required to use an additional complex adversarial training with an extra discriminator model. In this paper, we present RealUID, a universal distillation framework for all matching models that seamlessly incorporates real data into the distillation procedure without GANs. Our RealUID approach offers a simple theoretical foundation that covers previous distillation methods for Flow Matching and Diffusion models, and is also extended to their modifications, such as Bridge Matching and Stochastic Interpolants. The code can be found in https://github.com/David-cripto/RealUID.

翻译：尽管现代扩散模型、流模型及其他匹配模型在生成质量上表现卓越，但其推理速度缓慢，需要多步迭代生成。近期的蒸馏方法通过在前训练教师模型指导下训练高效的一步生成器来解决这一问题。然而，这些方法通常局限于单一特定框架，例如仅适用于扩散模型或仅适用于流模型。此外，这些方法本质上是无数据依赖的，若要利用真实数据，则需引入额外的复杂对抗训练及判别器模型。本文提出RealUID——一种适用于所有匹配模型的通用蒸馏框架，无需GAN即可将真实数据无缝整合至蒸馏过程中。我们的RealUID方法提供了简洁的理论基础，涵盖先前针对流匹配与扩散模型的蒸馏方法，并可扩展至其变体，如桥匹配与随机插值模型。代码可见于https://github.com/David-cripto/RealUID。