Large Language Model Steering of Acousto-Optic Tweezer Arrays
Abstract
Controlling quantum hardware requires choosing and revising experimental strategies as nonlinear response, coupling, and drift limit calibration. We experimentally investigate a general-purpose large language model (LLM) as a human-in-the-loop strategy layer for a neutral-atom tweezer platform. Through step-by-step interaction with experimental feedback, the LLM guides a transition from multiplicative row--column correction to a drift-corrected local response-matrix method when separable feedback stalls. Deterministic software performs image analysis and numerical updates, while an experimentalist verifies and authorizes each waveform. For a 10×10 crossed-acousto-optic-deflector array, this workflow reduces the Gaussian-peak coefficient of variation from 20.30% to 1.57% in 30 candidate-update rounds. The experiment demonstrates feedback-informed strategy selection on a physical quantum-control subsystem, rather than direct generation of control parameters alone. These results highlight the potential of LLMs to guide control optimization across diverse quantum platforms by adaptively formulating and refining experimental strategies.
Related
- Probing micron-range fifth forces with cold atoms in optical tweezer arrays
- Resonance-enhanced integrated acousto-optic beam steering.
- Focusing through coupled multimode fibers for miniaturized single-atom tweezer arrays
- A low-cost modular FPGA-based control system for neutral atom tweezer arrays
- Local polarization control of optical tweezer arrays
BibTeX
@misc{sheng2026large,
title = {{Large Language Model Steering of Acousto-Optic Tweezer Arrays}},
author = {Cheng Sheng and Kai-xuan Xiao and Yin Zhang and Wen-hao Wang and Feng-yi Zhang and Xu-xing Geng and Guang-ming Huang and Gao-xiang Li},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.25014},
eprint = {2610.25014},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.25014},
}
RIS
TY - UNPB
TI - Large Language Model Steering of Acousto-Optic Tweezer Arrays
AU - Cheng Sheng
AU - Kai-xuan Xiao
AU - Yin Zhang
AU - Wen-hao Wang
AU - Feng-yi Zhang
AU - Xu-xing Geng
AU - Guang-ming Huang
AU - Gao-xiang Li
PY - 2026
DA - 2026/10/18
PB - arXiv
DO - 10.48550/arxiv.2610.25014
UR - https://arxiv.org/abs/2610.25014
AB - Controlling quantum hardware requires choosing and revising experimental strategies as nonlinear response, coupling, and drift limit calibration. We experimentally investigate a general-purpose large language model (LLM) as a human-in-the-loop strategy layer for a neutral-atom tweezer platform. Through step-by-step interaction with experimental feedback, the LLM guides a transition from multiplicative row--column correction to a drift-corrected local response-matrix method when separable feedback stalls. Deterministic software performs image analysis and numerical updates, while an experimentalist verifies and authorizes each waveform. For a 10×10 crossed-acousto-optic-deflector array, this workflow reduces the Gaussian-peak coefficient of variation from 20.30% to 1.57% in 30 candidate-update rounds. The experiment demonstrates feedback-informed strategy selection on a physical quantum-control subsystem, rather than direct generation of control parameters alone. These results highlight the potential of LLMs to guide control optimization across diverse quantum platforms by adaptively formulating and refining experimental strategies.
ER -