Toward Greater Autonomy in Materials Discovery Agents: Unifying Planning, Physics, and Scientists

2025/06/05 by Zhou, Lianhao, Hongyi Ling, Keqiang Yan +12 · 1 citation
Decision Sciences · Materials Science · #Artificial Intelligence (cs.AI) #Computational Physics (physics.comp-ph) #Electronic and Structural Properties of Oxides #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning in Materials Science #Materials Science (cond-mat.mtrl-sci) #Scientific Computing and Data Management

paper · pdf · doi:10.48550/arxiv.2506.05616

openalex publication_date 2025/06/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We aim at designing language agents with greater autonomy for crystal materials discovery. While most of existing studies restrict the agents to perform specific tasks within predefined workflows, we aim to automate workflow planning given high-level goals and scientist intuition. To this end, we propose Materials Agent unifying Planning, Physics, and Scientists, known as MAPPS. MAPPS consists of a Workflow Planner, a Tool Code Generator, and a Scientific Mediator. The Workflow Planner uses large language models (LLMs) to generate structured and multi-step workflows. The Tool Code Generator synthesizes executable Python code for various tasks, including invoking a force field foundation model that encodes physics. The Scientific Mediator coordinates communications, facilitates scientist feedback, and ensures robustness through error reflection and recovery. By unifying planning, physics, and scientists, MAPPS enables flexible and reliable materials discovery with greater autonomy, achieving a five-fold improvement in stability, uniqueness, and novelty rates compared with prior generative models when evaluated on the MP-20 data. We provide extensive experiments across diverse tasks to show that MAPPS is a promising framework for autonomous materials discovery.

Citations

• MatLLMSearch: Crystal Structure Discovery with Evolution-Guided Large Language Models
• Invariant Tokenization of Crystalline Materials for Language Model Enabled Generation
• DrugAgent: Automating AI-aided Drug Discovery Programming through LLM Multi-Agent Collaboration
• FlowLLM: Flow Matching for Material Generation with Large Language Models as Base Distributions
• LLMatDesign: Autonomous Materials Discovery with Large Language Models
• A Comprehensive Survey of Scientific Large Language Models and Their Applications in Scientific Discovery
• A Space Group Symmetry Informed Network for O(3) Equivariant Crystal Tensor Prediction
• CRISPR-GPT for Agentic Automation of Gene-editing Experiments
• Complete and Efficient Graph Transformers for Crystal Material Property Prediction
• Toward a Team of AI-made Scientists for Scientific Discovery from Gene Expression Data
• Fine-Tuned Language Models Generate Stable Inorganic Materials as Text
• ProtAgents: Protein discovery via large language model multi-agent collaborations combining physics and machine learning
• Matbench Discovery -- A framework to evaluate machine learning crystal stability predictions
• ChatMOF: An Autonomous AI System for Predicting and Generating Metal-Organic Frameworks
• Crystal Structure Prediction by Joint Equivariant Diffusion
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Crystal Structure Generation with Autoregressive Large Language Modeling
• Periodic Graph Transformers for Crystal Material Property Prediction
• A universal graph deep learning interatomic potential for the periodic table
• Crystal Diffusion Variational Autoencoder for Periodic Material Generation
• Generalized Gradient Approximation Made Simple
• Efficient iterative schemes forab initiototal-energy calculations using a plane-wave basis set
• Self-Consistent Equations Including Exchange and Correlation Effects

Cited by

• AutoLabs: Cognitive Multi-Agent Systems with Self-Correction for Autonomous Chemical Experimentation
• CodePDE: An Inference Framework for LLM-driven PDE Solver Generation

Related

• VASP Agent: An Agentic Framework for Autonomous First-principles Calculations
• Large language models in materials science and the need for open-source approaches
• PriM: Principle-Inspired Material Discovery through Multi-Agent Collaboration
• LLM Agents for Knowledge Discovery in Atomic Layer Processing
• Hierarchical Multi-agent Large Language Model Reasoning for Autonomous Functional Materials Discovery