Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.05421 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916885307588608 |
|---|---|
| author | Sha, Rong Wang, Binglin Yang, Jun Ma, Xiaoxiao Wu, Chengkun Yan, Liang Zhou, Chao Liu, Jixun Wang, Guochao Yan, Shuhua Zhu, Lingxiao |
| author_facet | Sha, Rong Wang, Binglin Yang, Jun Ma, Xiaoxiao Wu, Chengkun Yan, Liang Zhou, Chao Liu, Jixun Wang, Guochao Yan, Shuhua Zhu, Lingxiao |
| contents | Large language models (LLM) exhibit broad utility but face limitations in quantum sensor development, stemming from interdisciplinary knowledge barriers and involving complex optimization processes. Here we present QCopilot, an LLM-based multi-agent framework integrating external knowledge access, active learning, and uncertainty quantification for quantum sensor design and diagnosis. Comprising commercial LLMs with few-shot prompt engineering and vector knowledge base, QCopilot employs specialized agents to adaptively select optimization methods, automate modeling analysis, and independently perform problem diagnosis. Applying QCopilot to atom cooling experiments, we generated 10${}^{\rm{8}}$ sub-$\rmμ$K atoms without any human intervention within a few hours, representing $\sim$100$\times$ speedup over manual experimentation. Notably, by continuously accumulating prior knowledge and enabling dynamic modeling, QCopilot can autonomously identify anomalous parameters in multi-parameter experimental settings. Our work reduces barriers to large-scale quantum sensor deployment and readily extends to other quantum information systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_05421 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LLM-based Multi-Agent Copilot for Quantum Sensor Sha, Rong Wang, Binglin Yang, Jun Ma, Xiaoxiao Wu, Chengkun Yan, Liang Zhou, Chao Liu, Jixun Wang, Guochao Yan, Shuhua Zhu, Lingxiao Quantum Physics Artificial Intelligence Atomic Physics Large language models (LLM) exhibit broad utility but face limitations in quantum sensor development, stemming from interdisciplinary knowledge barriers and involving complex optimization processes. Here we present QCopilot, an LLM-based multi-agent framework integrating external knowledge access, active learning, and uncertainty quantification for quantum sensor design and diagnosis. Comprising commercial LLMs with few-shot prompt engineering and vector knowledge base, QCopilot employs specialized agents to adaptively select optimization methods, automate modeling analysis, and independently perform problem diagnosis. Applying QCopilot to atom cooling experiments, we generated 10${}^{\rm{8}}$ sub-$\rmμ$K atoms without any human intervention within a few hours, representing $\sim$100$\times$ speedup over manual experimentation. Notably, by continuously accumulating prior knowledge and enabling dynamic modeling, QCopilot can autonomously identify anomalous parameters in multi-parameter experimental settings. Our work reduces barriers to large-scale quantum sensor deployment and readily extends to other quantum information systems. |
| title | LLM-based Multi-Agent Copilot for Quantum Sensor |
| topic | Quantum Physics Artificial Intelligence Atomic Physics |
| url | https://arxiv.org/abs/2508.05421 |