Hierarchical Multi-agent Large Language Model Reasoning for Autonomous Functional Materials Discovery

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rothfarb, Samuel, Davis, Megan C., Matanovic, Ivana, Li, Baikun, Holby, Edward F., Kort-Kamp, Wilton J. M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909964717522944
author Rothfarb, Samuel
Davis, Megan C.
Matanovic, Ivana
Li, Baikun
Holby, Edward F.
Kort-Kamp, Wilton J. M.
author_facet Rothfarb, Samuel
Davis, Megan C.
Matanovic, Ivana
Li, Baikun
Holby, Edward F.
Kort-Kamp, Wilton J. M.
contents Artificial intelligence is reshaping scientific exploration, but most methods automate procedural tasks without engaging in scientific reasoning, limiting autonomy in discovery. We introduce Materials Agents for Simulation and Theory in Electronic-structure Reasoning (MASTER), an active learning framework where large language models autonomously design, execute, and interpret atomistic simulations. In MASTER, a multimodal system translates natural language into density functional theory workflows, while higher-level reasoning agents guide discovery through a hierarchy of strategies, including a single agent baseline and three multi-agent approaches: peer review, triage-ranking, and triage-forms. Across two chemical applications, CO adsorption on Cu-surface transition metal (M) adatoms and on M-N-C catalysts, reasoning-driven exploration reduces required atomistic simulations by up to 90% relative to trial-and-error selection. Reasoning trajectories reveal chemically grounded decisions that cannot be explained by stochastic sampling or semantic bias. Altogether, multi-agent collaboration accelerates materials discovery and marks a new paradigm for autonomous scientific exploration.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13930
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hierarchical Multi-agent Large Language Model Reasoning for Autonomous Functional Materials Discovery
Rothfarb, Samuel
Davis, Megan C.
Matanovic, Ivana
Li, Baikun
Holby, Edward F.
Kort-Kamp, Wilton J. M.
Materials Science
Artificial Intelligence
Computation and Language
Machine Learning
Multiagent Systems
Artificial intelligence is reshaping scientific exploration, but most methods automate procedural tasks without engaging in scientific reasoning, limiting autonomy in discovery. We introduce Materials Agents for Simulation and Theory in Electronic-structure Reasoning (MASTER), an active learning framework where large language models autonomously design, execute, and interpret atomistic simulations. In MASTER, a multimodal system translates natural language into density functional theory workflows, while higher-level reasoning agents guide discovery through a hierarchy of strategies, including a single agent baseline and three multi-agent approaches: peer review, triage-ranking, and triage-forms. Across two chemical applications, CO adsorption on Cu-surface transition metal (M) adatoms and on M-N-C catalysts, reasoning-driven exploration reduces required atomistic simulations by up to 90% relative to trial-and-error selection. Reasoning trajectories reveal chemically grounded decisions that cannot be explained by stochastic sampling or semantic bias. Altogether, multi-agent collaboration accelerates materials discovery and marks a new paradigm for autonomous scientific exploration.
title Hierarchical Multi-agent Large Language Model Reasoning for Autonomous Functional Materials Discovery
topic Materials Science
Artificial Intelligence
Computation and Language
Machine Learning
Multiagent Systems
url https://arxiv.org/abs/2512.13930