Atomistic Generative Diffusion for Materials Modeling

TL;DR

This paper introduces AGeDi, a novel generative diffusion framework for atomistic systems that models atomic positions and types, enabling the creation of diverse, physically plausible atomic structures across different materials.

Contribution

It combines score-based diffusion for atomic positions with a new discrete diffusion process for atomic types, advancing atomistic generative modeling.

Findings

High fidelity and diversity in generated structures

Successful atomic type interpolation for bimetallic clusters

Steering sampling toward specific crystallographic symmetries

Abstract

We present a generative modeling framework for atomistic systems that combines score-based diffusion for atomic positions with a novel continuous-time discrete diffusion process for atomic types. This approach enables flexible and physically grounded generation of atomic structures across chemical and structural domains. Applied to metallic clusters and two-dimensional materials using the QCD and C2DB datasets, our models achieve strong performance in fidelity and diversity, evaluated using precision-recall metrics against synthetic baselines. We demonstrate atomic type interpolation for generating bimetallic clusters beyond the training distribution, and use classifier-free guidance to steer sampling toward specific crystallographic symmetries in two-dimensional materials. These capabilities are implemented in Atomistic Generative Diffusion (AGeDi), an open-source, extensible software package for atomistic generative diffusion modeling.