Denoising Diffusion Probabilistic Models, Jonathan Ho, Ajay Jain, Pieter Abbeel, 2020Advances in Neural Information Processing Systems, Vol. 33DOI: 10.48550/arXiv.2006.11239 - The original paper that introduced Denoising Diffusion Probabilistic Models, detailing the forward noising process, its Gaussian properties, and the derivation of the closed-form marginal distribution.
Generative Modeling by Estimating Gradients of the Data Distribution, Yang Song, Stefano Ermon, 2019Advances in Neural Information Processing Systems, Vol. 32DOI: 10.48550/arXiv.1907.05600 - Introduces score-based generative modeling, a related framework that also relies on progressively adding noise and learning to reverse it. Provides context for the role of Gaussian noise in generative diffusion processes.
Score-Based Generative Modeling through Stochastic Differential Equations, Yang Song, Jascha Sohl-Dickstein, Diederik P. Kingma, Abhishek Kumar, Stefano Ermon, Ben Poole, 2021International Conference on Learning RepresentationsDOI: 10.48550/arXiv.2011.13456 - Unifies Denoising Diffusion Probabilistic Models and Score-Based Generative Models under a framework of Stochastic Differential Equations, providing a general understanding of the forward process.