Generative art with language+diffusion models

September 16, 2022 โ€” July 16, 2023

Figure 1

Generative art using modern diffusion-backed image generators. The name-brand models are DALL-E 2, stable diffusion, Midjourney etc, which are diffusion + transformer models.

CLIP presumably goes somewhere near here.

For audio stuff see music diffusion.

1 Adventures in DIY models

2 Punditry

3 Mechanics

4 UIs

5 Incoming

6 References

Dhariwal, and Nichol. 2021. โ€œDiffusion Models Beat GANs on Image Synthesis.โ€ arXiv:2105.05233 [Cs, Stat].
Dutordoir, Saul, Ghahramani, et al. 2022. โ€œNeural Diffusion Processes.โ€
Han, Zheng, and Zhou. 2022. โ€œCARD: Classification and Regression Diffusion Models.โ€
Ho, Jain, and Abbeel. 2020. โ€œDenoising Diffusion Probabilistic Models.โ€ arXiv:2006.11239 [Cs, Stat].
Hoogeboom, Gritsenko, Bastings, et al. 2021. โ€œAutoregressive Diffusion Models.โ€ arXiv:2110.02037 [Cs, Stat].
Nichol, and Dhariwal. 2021. โ€œImproved Denoising Diffusion Probabilistic Models.โ€ arXiv:2102.09672 [Cs, Stat].
Sohl-Dickstein, Weiss, Maheswaranathan, et al. 2015. โ€œDeep Unsupervised Learning Using Nonequilibrium Thermodynamics.โ€ arXiv:1503.03585 [Cond-Mat, q-Bio, Stat].
Song, Yang, and Ermon. 2020a. โ€œGenerative Modeling by Estimating Gradients of the Data Distribution.โ€ In Advances In Neural Information Processing Systems.
โ€”โ€”โ€”. 2020b. โ€œImproved Techniques for Training Score-Based Generative Models.โ€ In Advances In Neural Information Processing Systems.
Song, Jiaming, Meng, and Ermon. 2021. โ€œDenoising Diffusion Implicit Models.โ€ arXiv:2010.02502 [Cs].
von Platen, Patil, Lozhkov, et al. 2022. โ€œDiffusers: State-of-the-Art Diffusion Models.โ€
Yang, Zhang, Hong, et al. 2022. โ€œDiffusion Models: A Comprehensive Survey of Methods and Applications.โ€