MatterSim-MT: A multi-task foundation model for in silico materials characterization

TL;DR

MatterSim-MT is a multi-task foundation model trained on extensive first-principles data, enabling accurate, scalable in silico materials characterization and complex property simulations beyond traditional methods.

Contribution

The paper introduces MatterSim-MT, a multi-task foundation model that significantly improves materials property prediction and simulation scalability across diverse structures and conditions.

Findings

Accurately predicts pressure-dependent phonon splitting in SiC.

Demonstrates electric hysteresis in BaTiO3.

Models redox transitions in cathode materials.

Abstract

Accurate property characterization is a major bottleneck in materials design. While first-principles methods and task-specific machine-learning models have driven important progress, they remain fundamentally limited in scalability and generalizability across the vast space of structures and properties relevant to real-world materials design. We present MatterSim-MT, a multi-task foundation model for in silico materials simulation and property characterization. The model is pretrained on over 35 million first-principles-labeled structures covering 89 elements, temperatures up to 5000 K and pressures up to 1000 GPa, and is fine-tuned on various properties including Bader charges, magnetic moments, Born effective charges, and dielectric matrices. Out of the box, MatterSim-MT not only serves as a foundation model for predicting material structure, dynamics and thermodynamics, its multi-task architecture also enables a wide range of complex simulations that cannot be captured by potential energy surfaces alone. For example, we demonstrate pressure-dependent LO-TO phonon splitting in SiC with close agreement with experiment, electric hysteresis in ferroelectric BaTiO3, and the cationic-to-anionic redox transition during delithiation of a Li-rich cathode material. Finally, we show that MatterSim-MT scales well with more data and parameters, can be efficiently fine-tuned to higher levels of theory, and can be efficiently extended to new systems via active learning. Overall, we believe this approach provides a scalable route to accurate in silico materials characterization.