Abstract

Training of general-purpose machine learning interatomic potentials (MLIPs) relies on large datasets with properties usually computed with density functional theory (DFT). A prerequisite for accurate MLIPs is that the DFT data are well converged to minimize numerical errors. A possible symptom of errors in DFT force components is nonzero net force. Here, we consider net forces in datasets including SPICE, Transition1x, ANI-1x, ANI-1xbb, AIMNet2, QCML, and OMol25. Several of these datasets suffer from significant nonzero DFT net forces. We also quantify individual force component errors by comparison to recomputed forces using more reliable DFT settings at the same level of theory, and we find significant discrepancies in force components averaging from 1.7 meV/Å in the SPICE dataset to 33.2 meV/Å in the ANI-1x dataset. These findings underscore the importance of well converged DFT data as increasingly accurate MLIP architectures become available.

Affiliated Institutions

Related Publications

Publication Info

Year
2025
Type
article
Volume
163
Issue
22
Citations
0
Access
Closed

External Links

View on DOI.org

Social Impact

Altmetric
PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

0
OpenAlex

Cite This

Domantas Kuryla, Fabian Berger, Gábor Cśanyi et al. (2025). How accurate are DFT forces? Unexpectedly large uncertainties in molecular datasets. The Journal of Chemical Physics , 163 (22) . https://doi.org/10.1063/5.0296997

Identifiers

DOI
10.1063/5.0296997