EVE rethinks neural architecture by replacing scalar units with local variational probabilistic neurons.
March 17, 2026
Original Paper
Exploring the Dimensions of a Variational Neuron
arXiv · 2603.13849
The Takeaway
Shifts uncertainty modeling from global latent variables to the individual neuron level. This provides a new framework for local capacity control and temporal persistence within neural networks, making internal states more observable and controllable.
From the abstract
We introduce EVE (Elemental Variational Expanse), a variational distributional neuron formulated as a local probabilistic computational unit with an explicit prior, an amortized posterior, and unit-level variational regularization. In most modern architectures, uncertainty is modeled through global latent variables or parameter uncertainty, while the computational unit itself remains scalar. EVE instead relocates probabilistic structure to the neuron level, making it locally observable and contr