Modular large language model agents for multi-task computational materials science
Модульные агенты на основе больших языковых моделей для многозадачной вычислительной материаловедческой науки
2026-03-26
SCID: 54.1/9pz7ku5z
Discuss with AI
MatSciAgentcomputational materials sciencecrystal structure generationmolecular dynamics simulationmulti-agent framework
Figures from the paper
Abstract (AI)
The integration of large language models (LLMs) with domain-specific computational tools offers a promising pathway to streamline and enhance materials science workflows. This paper presents MatSciAgent, a multi-agent framework capable of performing key tasks such as materials data retrieval, continuum simulation, crystal structure generation, and molecular dynamics simulation. At the core of the framework is the master agent, which interprets the user’s natural language query, identifies the task type, and delegates it to a corresponding task-specific agent equipped with appropriate computational tools. Leveraging databases such as the Materials Project and MatWeb, the framework retrieves and summarizes materials data with grounded, factual responses—addressing a key limitation of vanilla LLMs. In cases where the target material is not found in existing databases, a generative task-specific agent can propose plausible crystal structures. For simulation tasks, dedicated agents extract relevant parameters from the user query to conduct continuum simulations (e.g., Cellular Automata and Monte Carlo Annealing) and atomistic simulations (e.g., Molecular Dynamics) using both established software and custom code. The modular design of these agents and their associated tools enables seamless extensibility, allowing the framework to evolve as new capabilities are integrated.
Key Findings
1
A generative agent proposes plausible crystal structures when target materials are absent from available databases.
2
A master agent interprets user queries, identifies task types, and delegates them to specialized tool-equipped agents.
3
Integration with Materials Project and MatWeb enables grounded, factual materials-data retrieval and summarization, addressing limitations of vanilla LLMs.
4
MatSciAgent is a modular multi-agent framework that coordinates materials data retrieval, crystal structure generation, continuum simulation, and molecular dynamics.
5
The framework achieved 100% parameter-extraction success across five runs and consistent materials extraction in 9 of 10 runs, demonstrating high stability.
Research Object
MatSciAgent, a modular multi-agent framework integrating large language models with domain-specific computational materials science tools
Research Subject
The framework’s automation, stability, accuracy, and extensibility in performing materials data retrieval, crystal structure generation, continuum simulation, and molecular dynamics simulation
Publication Details
Publication Date
2026-03-26
Journal
Publisher
ISSN
Cited by
31
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai7
LLaMA: Open and Efficient Foundation Language Models2023
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation2013
The atomic simulation environment—a Python library for working with atoms2017
The Open Quantum Materials Database (OQMD): assessing the accuracy of DFT formation energies2015
The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery2024
Autonomous chemical research with large language models2023
Augmenting large language models with chemistry tools2024