<ul><li>Diffusion models have shown excellent performance in generative modeling but suffer from slow sampling speed.</li><li>Discrete diffusion models face challenges in capturing dependencies between elements due to the computational cost of processing high-dimensional joint distributions.</li><li>The proposed method introduces 'mixture' models for discrete diffusion that can capture dimensional correlations while being scalable.</li><li>Experimental results demonstrate the effectiveness of the method in distilling pretrained discrete diffusion models in image and language domains.</li></ul>

Distillation of Discrete Diffusion through Dimensional Correlations

Discover more