The OpenLAM Challenges

TL;DR

The paper discusses the development of Large Atom Models (LAMs), emphasizing the creation of open benchmarks and datasets, including the OpenLAM Crystal Philately competition, to advance scientific computation and materials science.

Contribution

It introduces the OpenLAM initiative, establishing comprehensive benchmarks and a large dataset for evaluating LAMs across the periodic table.

Findings

Collected over 19.8 million valid structures for LAM evaluation.

Launched the OpenLAM convex hull dataset with 1 million structures.

Facilitated advancements in generative modeling and materials science applications.

Abstract

Inspired by the success of Large Language Models (LLMs), the development of Large Atom Models (LAMs) has gained significant momentum in scientific computation. Since 2022, the Deep Potential team has been actively pretraining LAMs and launched the OpenLAM Initiative to develop an open-source foundation model spanning the periodic table. A core objective is establishing comprehensive benchmarks for reliable LAM evaluation, addressing limitations in existing datasets. As a first step, the LAM Crystal Philately competition has collected over 19.8 million valid structures, including 1 million on the OpenLAM convex hull, driving advancements in generative modeling and materials science applications.