arXiv:2603.02346v2 Announce Type: replace-cross Abstract: We introduce Large Electron Model, a single neural network model that produces variational wavefunctions of interacting electrons over the entire Hamiltonian parameter manifold. Our model employs the Fermi Sets architecture, a universal representation of many-body fermionic wavefunctions, which is further conditioned on Hamiltonian parameter and particle number. For interacting electrons in a two-dimensional harmonic potential, a single trained model accurately predicts the ground state wavefunction while generalizing across unseen coup
Advances in AI, particularly sophisticated neural network architectures, are now enabling applications in fundamental physics simulations that were previously intractable, marking a new frontier for AI's scientific utility.
This development indicates a significant leap in AI's capacity for scientific discovery and predictive modeling, potentially accelerating material science, quantum computing, and energy research by efficiently predicting complex electron behaviors.
The ability to accurately predict ground state wavefunctions of interacting electrons with a single, generalizable neural network shifts the paradigm from computationally intensive simulations to AI-driven discovery in quantum chemistry and condensed matter physics.
- · Material scientists
- · Quantum computing researchers
- · Physics-based AI companies
- · Drug discovery platforms
- · Traditional high-performance computing (for specific simulation tasks)
- · R&D labs reliant solely on empirical methods
Rapid acceleration of new material discovery with tailor-made electronic properties.
Development of novel quantum computing architectures based on AI-predicted optimal electron interactions.
Reduced time and cost for developing advanced battery technologies, superconductors, and catalysts, impacting the global energy transition.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG