Grounding LLMs in Scientific Discovery via Embodied Actions

TL;DR

EmbodiedAct enhances scientific discovery with LLMs by integrating active embodied actions and perception, improving reliability and accuracy in complex simulations and modeling tasks.

Contribution

This paper introduces EmbodiedAct, a framework that transforms scientific software into embodied agents with a perception-execution loop, advancing LLM applications in scientific discovery.

Findings

Outperforms existing baselines in engineering design tasks

Achieves state-of-the-art reliability and stability in long-horizon simulations

Enhances accuracy in scientific modeling

Abstract

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.