较新（24.3）加速Diffusion模型推理的方法，附带参考文献

1.采用fast ODE solvers：

Karras, T., Aittala, M., Aila, T., Laine, S.: Elucidating the design space of diffusionbased generative models. In: Conference on Neural Information Processing Systems (NeurIPS) (2022)

Lu, C., Zhou, Y., Bao, F., Chen, J., Li, C., Zhu, J.: Dpm-solver: A fast ode solver for diffusion probabilistic model sampling in around 10 steps. Conference on Neural Information Processing Systems (NeurIPS) 35, 5775–5787 (2022) 

2.将原来的扩散模型作为教师，蒸馏到更快的少步学生网络

Meng, C., Rombach, R., Gao, R., Kingma, D., Ermon, S., Ho, J., Salimans, T.: On distillation of guided diffusion models. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2023)

Salimans, T., Ho, J.: Progressive distillation for fast sampling of diffusion models. In: International Conference on Learning Representations (ICLR) (2022)

3.一些采用一致性模型训练

Song, Y., Dhariwal, P., Chen, M., Sutskever, I.: Consistency models. In: International Conference on Machine Learning (ICML) (2023)

Luo, S., Tan, Y., Huang, L., Li, J., Zhao, H.: Latent consistency models: Synthesizing high-resolution images with few-step inference. arXiv preprint arXiv:2310.04378 (2023)

、对抗学习

Sauer, A., Lorenz, D., Blattmann, A., Rombach, R.: Adversarial diffusion distillation. arXiv preprint arXiv:2311.17042 (2023)

Xu, Y., Zhao, Y., Xiao, Z., Hou, T.: Ufogen: You forward once large scale text-toimage generation via diffusion gans. arXiv preprint arXiv:2311.09257 (2023)1

、变量得分蒸馏

Wang, Z., Lu, C., Wang, Y., Bao, F., Li, C., Su, H., Zhu, J.: Prolificdreamer: High-fidelity and diverse text-to-3d generation with variational score distillation. Conference on Neural Information Processing Systems (NeurIPS) 36 (2024)

Yin, T., Gharbi, M., Zhang, R., Shechtman, E., Durand, F., Freeman, W.T., Park, T.: One-step diffusion with distribution matching distillation. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2024)、

矫正流

Liu, X., Gong, C., Liu, Q.: Flow straight and fast: Learning to generate and transfer data with rectified flow. arXiv preprint arXiv:2209.03003 (2022)

Liu, X., Zhang, X., Ma, J., Peng, J., et al.: Instaflow: One step is enough for highquality diffusion-based text-to-image generation. In: International Conference on Learning Representations

或者它们的结合

Liu, X., Zhang, X., Ma, J., Peng, J., et al.: Instaflow: One step is enough for highquality diffusion-based text-to-image generation. In: International Conference on Learning Representations

4.改用GAN做生成

. Sauer, A., Karras, T., Laine, S., Geiger, A., Aila, T.: Stylegan-t: Unlocking the power of gans for fast large-scale text-to-image synthesis. In: International Conference on Machine Learning (ICML) (2023)

Kang, M., Zhu, J.Y., Zhang, R., Park, J., Shechtman, E., Paris, S., Park, T.: Scaling up gans for text-to-image synthesis. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2023)