More Like this

1. Diffusion-GAN: Training GANs with Diffusion

   Wang, Zhendong; Zheng, Huangjie; He, Pengcheng; Chen, Weizhu; Zhou, Mingyuan (May 2023, International Conference on Learning Representations)
2. Spatial movement with diffusion and memory-based self-diffusion and cross-diffusion

   https://doi.org/10.1016/j.jde.2021.10.021  

   Shi, Junping; Wang, Chuncheng; Wang, Hao (December 2021, Journal of Differential Equations)
3. Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion

   Christopher, Jacob K; Cardei, Michael; Bartoldson, Brian R; Kailkhura, Bhavya; Fioretto, Ferdinando (February 2025, Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL))
4. Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion

   https://doi.org/10.18653/v1/2025.naacl-long.601  

   Christopher, Jacob K; Bartoldson, Brian R; Ben-Nun, Tal; Cardei, Michael; Kailkhura, Bhavya; Fioretto, Ferdinando (January 2025, Association for Computational Linguistics (NAACL))
5. Beta Diffusion

   Zhou, Mingyuan; Chen, Tianqi; Wang, Zhendong; Zheng, Huangjie (December 2023, Neural Information Processing Systems)

   We introduce beta diffusion, a novel generative modeling method that integrates demasking and denoising to generate data within bounded ranges. Using scaled and shifted beta distributions, beta diffusion utilizes multiplicative transitions over time to create both forward and reverse diffusion processes, maintaining beta distributions in both the forward marginals and the reverse conditionals, given the data at any point in time. Unlike traditional diffusion-based generative models relying on additive Gaussian noise and reweighted evidence lower bounds (ELBOs), beta diffusion is multiplicative and optimized with KL-divergence upper bounds (KLUBs) derived from the convexity of the KL divergence. We demonstrate that the proposed KLUBs are more effective for optimizing beta diffusion compared to negative ELBOs, which can also be derived as the KLUBs of the same KL divergence with its two arguments swapped. The loss function of beta diffusion, expressed in terms of Bregman divergence, further supports the efficacy of KLUBs for optimization. Experimental results on both synthetic data and natural images demonstrate the unique capabilities of beta diffusion in generative modeling of range-bounded data and validate the effectiveness of KLUBs in optimizing diffusion models, thereby making them valuable additions to the family of diffusion-based generative models and the optimization techniques used to train them. 

   more » « less