A latent-space extrapolation grade built into graph atomic cluster expansion foundation potentials
Year: 2026
Type: preprint
Abstract: Foundation machine-learning interatomic potentials cover broad configurational and chemical spaces, but their reliability can vary across the atomic environments encountered during a simulation. Here we introduce the calibrated Mahalanobis (CALM) extrapolation grade $γ$, a piecewise differentiable per-atom quantity integrated into GRACE foundation models and evaluated alongside energies and forces in a single model pass. We define $γ$ from nearest-cluster Mahalanobis distances in latent feature space, setting $γ=1$ from the training-distance distribution separately for each element and cluster. Controlled tests show that a normalized random projection of the invariant many-body basis detects structural and chemical extrapolation. On different foundation datasets, OMat24 and SMAX, $γ$ correlates with atomic force errors and separates structures with different error distributions. The CALM grade adds percent-level computational cost, and its spatial gradient guides uncertainty-biased data collection toward configurations with larger absolute force errors.
Source: arXiv (Cornell University)
Institution: Ruhr University Bochum
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7215072282