MIND: AI Co-Scientist for Material Research

2026-04-15

AI Summary

This work addresses the limitation of current AI-driven scientific research systems, which are largely confined to textual reasoning and lack automated experimental validation. The authors propose a large language model–powered multi-agent framework that achieves, for the first time, a closed-loop automation encompassing hypothesis generation, in situ simulation-based experimentation, and debate-driven verification. The framework integrates a machine-learned interatomic potential (SevenNet-Omni) to enable efficient materials simulations and employs a multi-agent debate mechanism to enhance the reliability of scientific reasoning. Designed with a modular architecture, the system supports flexible extension to diverse research scenarios via a web interface. The code and a live demonstration have been open-sourced, offering the materials science community a reproducible and interactive AI research platform.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Agentic search
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Large language models (LLMs) have enabled agentic AI systems for scientific discovery, but most approaches remain limited to textbased reasoning without automated experimental verification. We propose MIND, an LLM-driven framework for automated hypothesis validation in materials research. MIND organizes the scientific discovery process into hypothesis refinement, experimentation, and debate-based validation within a multi-agent pipeline. For experimental verification, the system integrates Machine Learning Interatomic Potentials, particularly SevenNet-Omni, enabling scalable in-silico experiments. We also provide a web-based user interface for automated hypothesis testing. The modular design allows additional experimental modules to be integrated, making the framework adaptable to broader scientific workflows. The code is available at: https://github.com/IMMS-Ewha/MIND, and a demonstration video at: https://youtu.be/lqiFe1OQzN4.

Problem

• AI co-scientist
• automated hypothesis validation
• materials research
• experimental verification
• large language models

Innovation

• LLM-driven framework
• automated hypothesis validation
• multi-agent pipeline
• Machine Learning Interatomic Potentials
• in-silico experiments