Agentic multi-fidelity learning of quasiparticle and excitonic properties

2026-06-05

AI Summary

This work addresses the pervasive numerical instabilities in high-throughput GW–Bethe–Salpeter equation calculations for low-dimensional nanomaterials, which often distort quasiparticle bandgaps and exciton binding energies. To overcome this challenge, the authors propose an agent-guided multifidelity learning framework that employs a structural agent to diagnose numerical fragility, assign confidence weights, and integrate sparse high-fidelity reference data. By combining transfer learning with Gaussian process regression, the method corrects computational artifacts while preserving essential physics, such as strain dependence. This approach uniquely integrates explicit numerical diagnostics with surrogate modeling—eschewing direct interpolation of raw first-principles results—and substantially enhances the accuracy and reliability of excited-state properties. It effectively mitigates pathological issues like spurious spikes and zero-gap collapse, demonstrating broad applicability across diverse strongly quantum-confined systems.

Abstract

Many-body GW-Bethe-Salpeter equation calculations are essential for accurate simulations of electronic structure and optical properties in modern low-dimensional nanomaterials. However, these methods are computationally demanding and can exhibit localized numerical instabilities or convergence failures that are difficult to detect within high-throughput workflows. We introduce an agent-guided multi-fidelity framework for correcting GW-Bethe-Salpeter excited-state landscapes in strained MoS2-WS2 bilayers. Across stacking registries, strain branches and reciprocal-space samplings, the workflow identifies spike-like excursions, near-zero-gap collapse and cross-fidelity inconsistencies associated with fragile long-wavelength dielectric screening. A structural agent evaluates calculations by assigning confidence weights and selectively using a small number of high-accuracy reference points. Machine learning models then transfer information across related systems and apply Gaussian process corrections to recover improved quasiparticle gaps and exciton binding energies, with calibrated uncertainty estimates. The approach corrects numerically induced artifacts without erasing physical strain dependence and substantially improves agreement with higher-fidelity references relative to a no-agent baseline. These results show that reliable surrogate learning for excited-state materials requires explicit diagnosis of numerical fragility, not direct interpolation of raw first-principles data points. The proposed framework is readily transferable to other optoelectronic nanomaterials characterized by strong quantum confinement, such as quantum dots, nanoribbons, layered two-dimensional semiconductors, and hybrid perovskite nanostructures.

Problem

• multi-fidelity learning
• numerical instability
• GW-Bethe-Salpeter equation
• excited-state properties
• quasiparticle gaps

Innovation

• agent-guided learning
• multi-fidelity modeling
• GW-BSE
• numerical fragility diagnosis
• excited-state correction