Good Views Are Not Enough: Global Counterexamples for Gaussian-Regularized JEPAs
The paper presents globally certified counterexamples showing that good views alone do not guarantee scene invariance or uniqueness up to rotation at global optima of Gaussian-regularized JEPAs. In a four-state model, every global minimizer splits a scene and has an equal-value partner outside its rotation orbit; a 512-dimensional model shows pooled Epps-Pulley regularization can force retention of nuisance and positive cross-modal alignment error. The authors also report that a whitened prediction JEPA aligns better to the scene than next-token prediction but has 0.2872 nats/token higher linear-readout test NLL under matched readout budgets and 1% observation noise. A Bayes-risk decomposition separates information lost through the observation channel from readout difficulty.
It provides theoretical counterexamples and concrete evidence that Gaussian regularization in JEPAs can undermine the assumed identifiability and prediction benefits.
Comments & Questions (0)
No comments yet
Be the first to comment!