Towards Fully Automated Molecular Simulations: Multi-Agent Framework for Simulation Setup and Force Field Extraction
This work addresses the bottleneck of manual simulation setup for materials scientists, aiming to accelerate materials discovery through automation, though it is incremental as a first step toward fully autonomous systems.
The paper tackles the complexity of simulation setup and force field selection in automated characterization of porous materials by proposing a multi-agent framework using LLM-based agents to autonomously plan simulations, assemble force fields, execute them, and interpret results, with initial evaluations showing high correctness and reproducibility.
Automated characterization of porous materials has the potential to accelerate materials discovery, but it remains limited by the complexity of simulation setup and force field selection. We propose a multi-agent framework in which LLM-based agents can autonomously understand a characterization task, plan appropriate simulations, assemble relevant force fields, execute them and interpret their results to guide subsequent steps. As a first step toward this vision, we present a multi-agent system for literature-informed force field extraction and automated RASPA simulation setup. Initial evaluations demonstrate high correctness and reproducibility, highlighting this approach's potential to enable fully autonomous, scalable materials characterization.