Our paper Sparsity-Inducing Categorical Prior Improves Robustness of the Information Bottleneck (https://arxiv.org/pdf/2203.02592.pdf) has been accepted at AISTATS 2023 (https://aistats.org/aistats2023/) and publication in the proceedings. This is a joint work with Michigan state university and Ian Fischer from Google Research. We present a novel sparsity-inducing spike-slab categorical prior that uses sparsity as a mechanism to provide the flexibility that allows each data point to learn its own dimension distribution. In addition, it provides a mechanism for learning a joint distribution of the latent variable and the sparsity, and hence it can account for the complete uncertainty in the latent space.