Authors: Shanchuan Lin, Xin Xia, Yuxi Ren, Ceyuan Yang, Xuefeng Xiao, Lu Jiang
Abstract: The diffusion models are widely used for image and video generation, but
their iterative generation process is slow and expansive. While existing
distillation approaches have demonstrated the potential for one-step generation
in the image domain, they still suffer from significant quality degradation. In
this work, we propose Adversarial Post-Training (APT) against real data
following diffusion pre-training for one-step video generation. To improve the
training stability and quality, we introduce several improvements to the model
architecture and training procedures, along with an approximated R1
regularization objective. Empirically, our experiments show that our
adversarial post-trained model, Seaweed-APT, can generate 2-second, 1280×720,
24fps videos in real time using a single forward evaluation step. Additionally,
our model is capable of generating 1024px images in a single step, achieving
quality comparable to state-of-the-art methods.
Source: http://arxiv.org/abs/2501.08316v1
