对抗概念蒸馏用于一步扩散个性化
Adversarial Concept Distillation for One-Step Diffusion Personalization
AI总结：
本文提出OPAD框架，结合教师-学生蒸馏与对抗监督，通过多步扩散模型作为教师，一步学生模型联合训练，以实现一步扩散模型的高效个性化。
摘要：
最近在加速文本到图像扩散模型方面的进展使得在单步去噪步骤内能够生成高保真的图像。然而，定制快速的一步模型仍然具有挑战性，因为现有方法始终无法产生可接受的结果，凸显了需要新方法来个性化一步模型的必要性。因此，我们提出了一步个性化对抗蒸馏（OPAD），该框架结合了教师-学生蒸馏与对抗监督。多步扩散模型作为教师，而一步学生模型则与之共同训练。学生通过保持与教师一致的对齐损失以及对齐其输出与真实图像分布的对抗损失进行学习。除了一步个性化之外，我们还发现学生高效的生成和对抗增强的表示为改进教师模型提供了有价值的反馈，形成了一种协作学习阶段。广泛的实验表明，OPAD是首个能够为一步扩散模型提供可靠、高质量个性化的方法；相比之下，先前方法大多失败并产生严重的失败案例，而OPAD保持了单步效率。
英文摘要：
Recent progress in accelerating text-to-image diffusion models enables high-fidelity synthesis within a single denoising step. However, customizing the fast one-step models remains challenging, as existing methods consistently fail to produce acceptable results, underscoring the need for new methodologies to personalize one-step models. Therefore, we propose One-step Personalized Adversarial Distillation (OPAD), a framework that combines teacher-student distillation with adversarial supervision. A multi-step diffusion model serves as the teacher, while a one-step student model is jointly trained with it. The student learns from alignment losses that preserve consistency with the teacher and from adversarial losses that align its output with real image distributions. Beyond one-step personalization, we further observe that the student's efficient generation and adversarially enriched representations provide valuable feedback to improve the teacher model, forming a collaborative learning stage. Extensive experiments demonstrate that OPAD is the first approach to deliver reliable, high-quality personalization for one-step diffusion models; in contrast, prior methods largely fail and produce severe failure cases, while OPAD preserves single-step efficiency.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
