AI RESEARCH
Historical Consensus: Preventing Posterior Collapse via Iterative Selection of Gaussian Mixture Priors
arXiv CS.AI
•
ArXi:2603.10935v1 Announce Type: cross Variational autoencoders (VAEs) frequently suffer from posterior collapse, where latent variables become uninformative and the approximate posterior degenerates to the prior. Recent work has characterized this phenomenon as a phase transition governed by the spectral properties of the data covariance matrix.