Large Language Models for Physics Instrument Design
This work addresses automating instrument design for physics researchers, offering a hybrid approach that reduces human supervision, but it is incremental as it builds on existing RL methods.
The study tackled the problem of physics instrument design by comparing large language models (LLMs) to reinforcement learning (RL), finding that LLMs consistently generate valid and resource-aware configurations without task-specific training, though RL yields stronger final designs.
We study the use of large language models (LLMs) for physics instrument design and compare their performance to reinforcement learning (RL). Using only prompting, LLMs are given task constraints and summaries of prior high-scoring designs and propose complete detector configurations, which we evaluate with the same simulators and reward functions used in RL-based optimization. Although RL yields stronger final designs, we find that modern LLMs consistently generate valid, resource-aware, and physically meaningful configurations that draw on broad pretrained knowledge of detector design principles and particle--matter interactions, despite having no task-specific training. Based on this result, as a first step toward hybrid design workflows, we explore pairing the LLMs with a dedicated trust region optimizer, serving as a precursor to future pipelines in which LLMs propose and structure design hypotheses while RL performs reward-driven optimization. Based on these experiments, we argue that LLMs are well suited as meta-planners: they can design and orchestrate RL-based optimization studies, define search strategies, and coordinate multiple interacting components within a unified workflow. In doing so, they point toward automated, closed-loop instrument design in which much of the human effort required to structure and supervise optimization can be reduced.