# Grounding LLMs in Scientific Discovery via Embodied Actions

arXiv: https://arxiv.org/abs/2602.20639 · HTML: https://mlsanity.com/paper/2602.20639/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 37 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Builds on (most cited first, 27 of 27)

- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering](https://mlsanity.com/paper/2410.07095/) (2024)
- [Tool Learning with Large Language Models: A Survey](https://mlsanity.com/paper/2405.17935/) (2024)
- [Chain of Code: Reasoning with a Language Model-Augmented Code Emulator](https://mlsanity.com/paper/2312.04474/) (2023)
- [AutoSurvey: Large Language Models Can Automatically Write Surveys](https://mlsanity.com/paper/2406.10252/) (2024)
- [Large Language Models for Automated Open-domain Scientific Hypotheses Discovery](https://mlsanity.com/paper/2309.02726/) (2023)
- [SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoning](https://mlsanity.com/paper/2409.05556/) (2024)
- [BLADE: Benchmarking Language Model Agents for Data-Driven Science](https://mlsanity.com/paper/2408.09667/) (2024)
- [MOOSE-Chem: Large Language Models for Rediscovering Unseen Chemistry Scientific Hypotheses](https://mlsanity.com/paper/2410.07076/) (2024)
- [ResearchBench: Benchmarking LLMs in Scientific Discovery via Inspiration-Based Task Decomposition](https://mlsanity.com/paper/2503.21248/) (2025)
- [ProtAgents: Protein discovery via large language model multi-agent collaborations combining physics and machine learning](https://mlsanity.com/paper/2402.04268/) (2024)
- [The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs](https://mlsanity.com/paper/2509.09677/) (2025)
- [LitLLM: A Toolkit for Scientific Literature Review](https://mlsanity.com/paper/2402.01788/) (2024)
- [Nova: An Iterative Planning and Search Approach to Enhance Novelty and Diversity of LLM Generated Ideas](https://mlsanity.com/paper/2410.14255/) (2024)
- [CodeScientist: End-to-End Semi-Automated Scientific Discovery with Code-based Experimentation](https://mlsanity.com/paper/2503.22708/) (2025)
- [Towards Execution-Grounded Automated AI Research](https://mlsanity.com/paper/2601.14525/) (2026)
- [Tools Fail: Detecting Silent Errors in Faulty Tools](https://mlsanity.com/paper/2406.19228/) (2024)
- [Advancing Multi-Agent Systems Through Model Context Protocol: Architecture, Implementation, and Applications](https://mlsanity.com/paper/2504.21030/) (2025)
- [Improving Scientific Hypothesis Generation with Knowledge Grounded Large Language Models](https://mlsanity.com/paper/2411.02382/) (2024)
- [OmniScientist: Toward a Co-evolving Ecosystem of Human and AI Scientists](https://mlsanity.com/paper/2511.16931/) (2025)
- [Gravity-Bench-v1: A Benchmark on Gravitational Physics Discovery for Agents](https://mlsanity.com/paper/2501.18411/) (2025)
- [Harnessing the Power of Adversarial Prompting and Large Language Models for Robust Hypothesis Generation in Astronomy](https://mlsanity.com/paper/2306.11648/) (2023)
- [MAPS: Advancing Multi-Modal Reasoning in Expert-Level Physical Science](https://mlsanity.com/paper/2501.10768/) (2025)
- [SimuAgent: An LLM-Based Simulink Modeling Assistant Enhanced with Reinforcement Learning](https://mlsanity.com/paper/2601.05187/) (2026)
- [LDC: Learning to Generate Research Idea with Dynamic Control](https://mlsanity.com/paper/2412.14626/) (2024)

## Abstract (arXiv)

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.