Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows

2026-05-14

AI Summary

This work addresses the limitations of existing automated workflows for machine learning interatomic potentials (MLIPs), which often rely on fixed pipelines or expert knowledge and struggle to generalize across heterogeneous material systems. To overcome this, the authors propose a large language model (LLM)-driven multi-agent autonomous framework that formulates MLIP development as a sequential decision-making problem. Guided by natural language instructions, multiple agents dynamically select optimization actions, enabling end-to-end automatic construction, adaptive refinement, and backtracking without predefined procedural constraints. Experiments on complex solid–electrolyte interphase (SEI) systems demonstrate that the proposed framework substantially enhances adaptability to unseen materials and empowers non-experts to efficiently develop high-accuracy MLIPs.

Abstract

Developing machine learning interatomic potentials (MLIPs) for complex materials systems remains challenging because it requires expertise in atomistic simulations, machine learning, and workflow design, as well as iterative active learning procedures. Existing automated pipelines typically assume a fixed sequence of stages or depend on domain experts, which limits their adaptability to heterogeneous materials systems where the optimal curriculum is not known in advance. To lower the barrier to developing MLIPs for non-experts, we propose Lang2MLIP, a multi-agent framework that takes natural-language input and formulates end-to-end MLIP development as a sequential decision-making problem solved by large language models (LLMs). At each step, a decision-making agent observes the current dataset, model, evaluation results, and execution log, and then automatically selects an appropriate action to improve the model. This removes the need for a predefined pipeline and enables the agent to self-correct by revisiting earlier subsystems when new failures arise. We evaluate this approach on a solid electrolyte interphase (SEI) system with multiple components and interfaces. These results suggest that LLM-based multi-agent systems are a promising direction for automating MLIP development and making it more accessible to non-experts.

Problem

• machine learning interatomic potentials
• complex materials systems
• automated pipeline
• active learning
• heterogeneous materials

Innovation

• Lang2MLIP
• multi-agent framework
• large language models
• machine learning interatomic potentials
• autonomous workflow