FP-DeErr: Application-Oriented Error Decomposition for Foundation Potentials

Abstract

Foundation potentials (FPs) have emerged as a new basis for atomistic modeling. While their evaluation using average energy and force errors often indicates near-DFT accuracy, their performance in practical computational studies remains inconsistent. While recent benchmarks evaluate FPs on downstream computational tasks, the underlying errors that determine task success or failure are not always clear. Here, we develop FP-DeErr, an application-oriented error-decomposition framework that resolves FP errors according to the physically meaningful quantities, configurations, and computational stages governing specific tasks. Rather than averaging errors over an entire dataset, FP-DeErr uses decomposed error metrics devised for physically meaningful quantities, focusing on the configurations where errors arise and matter most. We demonstrate FP-DeErr by evaluating multiple state-of-the-art FPs on three fundamental tasks: force prediction for atomistic simulations, energy ranking for substitutional and vacancy orderings, and ion/vacancy migration. These error-decomposition metrics resolve force errors among highly accurate, large-error, and far-from-equilibrium atoms; relative-energy errors among competing orderings, phases, and compositions; and ion migration errors among endpoints and along-path errors. By identifying where FP errors arise, this error decomposition provides targeted guidance for FP development. FP-DeErr also provides an open benchmark, evaluation code, and a public leaderboard for rigorous FP assessment.

A latent-space extrapolation grade built into graph atomic cluster expansion foundation potentials
Yury Lysogorskiy, Anton Bochkarev, Ralf Drautz
arXiv30 Sep 2026PreprintCC BY 4.0Full textCondensed matterarXiv
Diagnosing Latent Energy Decomposition in Machine-Learning Interatomic Potentials via Interacting Quantum Atoms
Kohei Shimamura, Ken-ichi Nomura
arXiv1 Sep 2026PreprintarXiv licenceFull textPhysical chemistry Computational physicsarXiv
Foundation-Model Supervision of Distilled Potentials for Million-Atom Molecular Dynamics
Xiaoqing Liu, Yangshuai Wang, Lei Shen
arXiv16 Oct 2026PreprintarXiv licenceFull textComputational physicsarXiv
Mixture of Polyconvex Neural Potentials for Parametric Hyperelasticity: Towards Foundation Material Models
Steven J. Yang, Govinda Anantha Padmanabha, D. Thomas Seidl +1
arXiv1 Sep 2026PreprintCC BY 4.0Full textComputational physics Condensed matterarXiv
A practical route to amorphous solid-state electrolyte structures by hybrid reverse Monte Carlo with foundation potentials
Xintian Wang, Xinyu Zhang, Denys Butenko +3
arXiv14 Oct 2026PreprintCC BY 4.0Full textCondensed matterarXiv

Cite

BibTeX RIS

BibTeX

@misc{amirian2026fpdeerr,
  title = {{FP-DeErr: Application-Oriented Error Decomposition for Foundation Potentials}},
  author = {Kiyan Amirian and Ramanuja Srinivasan Saravanan and Felix Adams and Charles E Schwarz and Yifei Mo},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.05714},
  eprint = {2609.05714},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.05714},
}

RIS

TY  - UNPB
TI  - FP-DeErr: Application-Oriented Error Decomposition for Foundation Potentials
AU  - Kiyan Amirian
AU  - Ramanuja Srinivasan Saravanan
AU  - Felix Adams
AU  - Charles E Schwarz
AU  - Yifei Mo
PY  - 2026
DA  - 2026/09/29
PB  - arXiv
DO  - 10.48550/arxiv.2609.05714
UR  - https://arxiv.org/abs/2609.05714
AB  - Foundation potentials (FPs) have emerged as a new basis for atomistic modeling. While their evaluation using average energy and force errors often indicates near-DFT accuracy, their performance in practical computational studies remains inconsistent. While recent benchmarks evaluate FPs on downstream computational tasks, the underlying errors that determine task success or failure are not always clear. Here, we develop FP-DeErr, an application-oriented error-decomposition framework that resolves FP errors according to the physically meaningful quantities, configurations, and computational stages governing specific tasks. Rather than averaging errors over an entire dataset, FP-DeErr uses decomposed error metrics devised for physically meaningful quantities, focusing on the configurations where errors arise and matter most. We demonstrate FP-DeErr by evaluating multiple state-of-the-art FPs on three fundamental tasks: force prediction for atomistic simulations, energy ranking for substitutional and vacancy orderings, and ion/vacancy migration. These error-decomposition metrics resolve force errors among highly accurate, large-error, and far-from-equilibrium atoms; relative-energy errors among competing orderings, phases, and compositions; and ion migration errors among endpoints and along-path errors. By identifying where FP errors arise, this error decomposition provides targeted guidance for FP development. FP-DeErr also provides an open benchmark, evaluation code, and a public leaderboard for rigorous FP assessment.
ER  -

Abstract

Related

Cite