Elevating universal interatomic potentials with large-scale off-equilibrium data

Jiayu Peng

doi:10.1038/s43588-026-00994-y

Elevating universal interatomic potentials with large-scale off-equilibrium data

Jiayu Peng

Department of Materials Design and Innovation

Research output: Contribution to journal › Comment/debate

Abstract

The Open Materials 2024 dataset provides a large-scale, open-access collection of quantum-chemical atomistic simulations that encompasses diverse off-equilibrium crystal structures, thereby making machine-learning interatomic potentials more robust across different materials prediction tasks.

Original language	English
Pages (from-to)	552-553
Number of pages	2
Journal	Nature Computational Science
Volume	6
Issue number	6
DOIs	https://doi.org/10.1038/s43588-026-00994-y
State	Published - Jun 2026

Access to Document

10.1038/s43588-026-00994-y

Cite this

@article{7ec6f7369f7445e9afe4fcd7776135e4,

title = "Elevating universal interatomic potentials with large-scale off-equilibrium data",

abstract = "The Open Materials 2024 dataset provides a large-scale, open-access collection of quantum-chemical atomistic simulations that encompasses diverse off-equilibrium crystal structures, thereby making machine-learning interatomic potentials more robust across different materials prediction tasks.",

author = "Jiayu Peng",

note = "Publisher Copyright: {\textcopyright} Springer Nature America, Inc. 2026.",

year = "2026",

month = jun,

doi = "10.1038/s43588-026-00994-y",

language = "English",

volume = "6",

pages = "552--553",

journal = "Nature Computational Science",

issn = "2662-8457",

publisher = "Springer Nature",

number = "6",

}

TY - JOUR

T1 - Elevating universal interatomic potentials with large-scale off-equilibrium data

AU - Peng, Jiayu

N1 - Publisher Copyright: © Springer Nature America, Inc. 2026.

PY - 2026/6

Y1 - 2026/6

N2 - The Open Materials 2024 dataset provides a large-scale, open-access collection of quantum-chemical atomistic simulations that encompasses diverse off-equilibrium crystal structures, thereby making machine-learning interatomic potentials more robust across different materials prediction tasks.

AB - The Open Materials 2024 dataset provides a large-scale, open-access collection of quantum-chemical atomistic simulations that encompasses diverse off-equilibrium crystal structures, thereby making machine-learning interatomic potentials more robust across different materials prediction tasks.

UR - https://www.scopus.com/pages/publications/105041169033

U2 - 10.1038/s43588-026-00994-y

DO - 10.1038/s43588-026-00994-y

M3 - Comment/debate

AN - SCOPUS:105041169033

SN - 2662-8457

VL - 6

SP - 552

EP - 553

JO - Nature Computational Science

JF - Nature Computational Science

IS - 6

ER -

Elevating universal interatomic potentials with large-scale off-equilibrium data

Abstract

Access to Document

Other files and links

Fingerprint

Cite this