EVE is a neural quantum state that enforces exact momentum eigenstates by construction, allowing VMC to variationally solve quasiparticle states across multiple phases in 2D interacting bosons.
Better, faster fermionic neural networks
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
Basis-free neural-network geminal and Jastrow factors inside an AGP ansatz achieve sub-millihartree accuracy for H2 and rectangular H4 in VMC while exposing nodal errors at square H4 geometry.
Neural wave functions uncover FFLO, polarized superfluid, phase-separated, and crystalline Cooper-pair phases in the 2D spin-imbalanced Fermi gas.
The paper introduces neural-network trial wave functions for variational Monte Carlo, frames the variational method as unsupervised learning, and illustrates the approach on the Yukawa potential and hydrogen molecule.