arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.20639cs.AI
收藏

通过具身动作 grounding LLMs 在科学发现中

Grounding LLMs in Scientific Discovery via Embodied Actions

Tsinghua University, Beijing, China(清华大学)

Bo Zhang, Jinfeng Zhou, Yuxuan Chen, Jianing Yin, Minlie Huang, Hongning Wang

AI总结:

通过具身动作框架,LLMs在科学发现中实现更稳定的物理模拟和更高的建模准确性。

摘要:

大型语言模型(LLMs)在科学发现中展现出显著潜力,但难以弥合理论推理与可验证的物理模拟之间的差距。现有解决方案在被动的“执行-然后响应”循环中运行,因此缺乏运行时感知,使代理难以察觉瞬时异常(例如数值不稳定性或发散振荡)。为了解决这一限制,我们提出了EmbodiedAct框架,通过将LLMs grounding在具身动作中,将已有的科学软件转化为主动的具身代理。我们将在MATLAB中实例化EmbodiedAct,并在复杂的工程设计和科学建模任务上进行评估。大量实验表明,EmbodiedAct在现有基线中表现显著优于现有方法,通过确保长时间模拟中的满意可靠性和稳定性以及科学建模中的增强准确性,实现了SOTA性能。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑