SDXL-Lightning：漸進式對抗擴散蒸餾

SDXL-Lightning: Progressive Adversarial Diffusion Distillation

February 21, 2024

作者: Shanchuan Lin, Anran Wang, Xiao Yang

cs.AI

摘要

我們提出了一種擴散蒸餾方法，基於SDXL，在一步/幾步 1024px 文本到圖像生成中實現了新的最先進技術。我們的方法結合了漸進式和對抗式蒸餾，以實現質量和模式覆蓋之間的平衡。在本文中，我們討論了理論分析、鑑別器設計、模型公式和訓練技術。我們將我們蒸餾的SDXL-Lightning模型以LoRA和完整的UNet權重的形式開源。

English

We propose a diffusion distillation method that achieves new state-of-the-art in one-step/few-step 1024px text-to-image generation based on SDXL. Our method combines progressive and adversarial distillation to achieve a balance between quality and mode coverage. In this paper, we discuss the theoretical analysis, discriminator design, model formulation, and training techniques. We open-source our distilled SDXL-Lightning models both as LoRA and full UNet weights.