VJE is a new variational non-contrastive SSL method that models target embeddings with a directional-radial Student-t distribution to enable structured uncertainty estimation directly in the learned representation space.
Over the same training run,k–NN accuracy on the encoder outputzdeclines from 8.7% to 6.3%, whilek–NN on the posterior meanµremains at chance level throughout
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Joint Embedding Variational Bayes
VJE is a new variational non-contrastive SSL method that models target embeddings with a directional-radial Student-t distribution to enable structured uncertainty estimation directly in the learned representation space.