让大模型通过实时感知与操作，实现更可靠的科学仿真。
Grounding LLMs in Scientific Discovery via Embodied Actions
- 将大模型嵌入物理仿真系统，实现感知-执行闭环控制。
- 在长周期仿真中保持稳定，准确率显著优于现有方法。
- 适合需要高可靠性的工程设计与科学建模场景。
大语言模型在科学发现中展现出巨大潜力，但在理论推理与可验证的物理仿真之间仍存在鸿沟。现有方法采用被动的‘执行-响应’循环，缺乏运行时感知能力，难以捕捉瞬态异常（如数值不稳定或发散振荡）。为此，我们提出EmbodiedAct框架，通过紧密的感知-执行回路，将成熟科学软件转化为具备身体性行为的主动智能体。我们在MATLAB中实现了该框架，并在复杂工程设计与科学建模任务上进行了评估。大量实验表明，EmbodiedAct显著优于现有基线，在长周期仿真中确保了更高的可靠性与稳定性，同时提升了科学建模的准确性，达到当前最优水平。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have shown significant potential in scientific discovery but struggle to bridge the gap between theoretical reasoning and verifiable physical simulation. Existing solutions operate in a passive "execute-then-response" loop and thus lacks runtime perception, obscuring agents to transient anomalies (e.g., numerical instability or diverging oscillations). To address this limitation, we propose EmbodiedAct, a framework that transforms established scientific software into active embodied agents by grounding LLMs in embodied actions with a tight perception-execution loop. We instantiate EmbodiedAct within MATLAB and evaluate it on complex engineering design and scientific modeling tasks. Extensive experiments show that EmbodiedAct significantly outperforms existing baselines, achieving SOTA performance by ensuring satisfactory reliability and stability in long-horizon simulations and enhanced accuracy in scientific modeling.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。