Work

Convolutional transformer wave functions

Year: 2026

Type: article

Abstract: Deep neural quantum states have recently achieved remarkable performance in solving challenging quantum many-body problems. While transformer networks appear particularly promising due to their success in computer science, we show that previously reported transformer wave functions have not so far been capable of utilizing their full power. Here, we introduce the convolutional transformer wave function (CTWF). We show that our CTWF exhibits state-of-the-art performance in ground-state search and nonequilibrium dynamics compared to previous results, demonstrating its promising capacity in complex quantum problems.

Source: Physical Review Research

DOI: 10.1103/7xwp-25y9

Authors: Ao Chen, Vighnesh Dattatraya Naik, Markus Heyl

Institutions: University of Augsburg, Flatiron Institute

Language: English

Topic: Quantum many-body systems

Subfield: Atomic and Molecular Physics, and Optics

Field: Physics and Astronomy

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,625

Funders: Deutsche Forschungsgemeinschaft, Horizon 2020