machine-learning-interatomic-potentials

The article explains how NVIDIA ALCHEMI Toolkit uses PyTorch-native building blocks to construct GPU-accelerated materials simulation workflows with AI coding agents, detailing an end-to-end process and providing practical guidance validated on NVIDIA H200 GPUs.

This paper proposes using atom-averaged features from pretrained MLIPs like MACE as coarse coordinates for evaluating and guiding inorganic crystal structure generation, introducing the Coarse-Fine Transport Distance (CFTD) metric that captures both quality and novelty in a distribution-based framework.