Discovering Models for Experimental Quantum Systems with Agentic AI
Abstract
A central obstacle in realizing color center cavity-QED-based quantum technologies is inhomogeneity, or disorder, among individual emitters. As the number of emitters increases, so does the complexity of the experimental data, making it difficult to quantify the present physical mechanisms of disorder. Here we introduce an artificial intelligence-driven model discovery framework in which a large language model (LLM)-based agent, equipped with tools and a structured workflow, proposes candidate Hamiltonian and Lindbladian models, generates computational tools, and quantitatively evaluates competing physical hypotheses. We apply the workflow to three previously published experimental cavity-QED datasets collected from color centers in diamond and silicon carbide. Across the experiments, the agent successfully identifies physical models for the system and characterizes the disorder. In assessing reproducibility, we find that run-to-run variability is concentrated primarily in the research plan generation step, while the computational analysis itself is substantially more consistent. Our results suggest that structured LLM agent workflows can support model discovery from quantum optical experimental data while simultaneously allowing for human expert supervision and collaboration.
Related
- Discovering state-of-the-art quantum circuits with large language models
- Discovering New Problems for Decoded Quantum Interferometry
- Sparse-QITE: Discovering Compact Generators for Quantum Imaginary-Time Evolution via Regularization Methods
- ERAF4XRD: A multimodal agentic framework for constructing validated experimental X-ray diffraction databases from scientific literature
- Principles and Techniques for Agentic Research in Experimental Physics
BibTeX
@misc{kleidermacher2026discovering,
title = {{Discovering Models for Experimental Quantum Systems with Agentic AI}},
author = {Hannah C. Kleidermacher and Bennet Windt and Miguel Bello and James Zou and Jelena Vučković},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.00620},
eprint = {2611.00620},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.00620},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Discovering Models for Experimental Quantum Systems with Agentic AI
AU - Hannah C. Kleidermacher
AU - Bennet Windt
AU - Miguel Bello
AU - James Zou
AU - Jelena Vučković
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.00620
UR - https://arxiv.org/abs/2611.00620
AB - A central obstacle in realizing color center cavity-QED-based quantum technologies is inhomogeneity, or disorder, among individual emitters. As the number of emitters increases, so does the complexity of the experimental data, making it difficult to quantify the present physical mechanisms of disorder. Here we introduce an artificial intelligence-driven model discovery framework in which a large language model (LLM)-based agent, equipped with tools and a structured workflow, proposes candidate Hamiltonian and Lindbladian models, generates computational tools, and quantitatively evaluates competing physical hypotheses. We apply the workflow to three previously published experimental cavity-QED datasets collected from color centers in diamond and silicon carbide. Across the experiments, the agent successfully identifies physical models for the system and characterizes the disorder. In assessing reproducibility, we find that run-to-run variability is concentrated primarily in the research plan generation step, while the computational analysis itself is substantially more consistent. Our results suggest that structured LLM agent workflows can support model discovery from quantum optical experimental data while simultaneously allowing for human expert supervision and collaboration.
ER -