Heavy-Tailed Class-Conditional Priors for Long-Tailed Generative Modeling

ArXi:2509.02154v2 Announce Type: replace-cross Variational Autoencoders (VAEs) with global priors trained under an imbalanced empirical class distribution can lead to underrepresentation of tail classes in the latent space. While $t^3$VAE improves robustness via heavy-tailed Student's $t$-distribution priors, its single global prior still allocates mass proportionally to class frequency. We address this latent geometric bias by