Necessary conditions for benchmarking next-generation AI in materials discovery
Abstract
Current AI predicts materials by speeding up energy minimisation and training on large databases. However, scaling data generation prevents manual analysis by domain experts. If novelty is evaluated by chemical compositions or traditional representations in Crystallographic Information Files, or standard structure matchers based on manually chosen thresholds, one can easily generate millions of ``new'' compositions or files without discovering truly novel materials. This paper uses a hierarchy of crystal invariants to continuously trace and quantify structural novelty based on the new mathematical definition of a crystal structure and asymptotic guarantees for invariant descriptors. The invariant-based analysis through cascade computations on a modest computer exposes the unexpected scale of duplication in the world's largest materials databases, shown here in the high-profile case of GNoME datasets. To prevent current ``embarrassment of over-prediction'', we propose necessary conditions for next-generation AI that should be automatically enforced before claiming any new materials.
Related
- Necessary and sufficient conditions for optimal control of coupled forward-backward stochastic control systems driven by G-Brownian motion
- Higher-order necessary conditions for nonscattering
- Necessary and Sufficient Conditions for Adaptive Stabilizability of Continuous-Time Markov Jump Linear Stochastic Systems
- Necessary Optimality Conditions for Bilevel Programming using Tangential Subdifferentials
- Necessary and sufficient conditions for the self-normalized central limit theorem for independent triangular arrays
BibTeX
@misc{widdowson2026necessary,
title = {{Necessary conditions for benchmarking next-generation AI in materials discovery}},
author = {Daniel E Widdowson and Tatiana V Kurlina and Olga D Anosova and Vitaliy A Kurlin},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.13885},
eprint = {2610.13885},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.13885},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Necessary conditions for benchmarking next-generation AI in materials discovery
AU - Daniel E Widdowson
AU - Tatiana V Kurlina
AU - Olga D Anosova
AU - Vitaliy A Kurlin
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2610.13885
UR - https://arxiv.org/abs/2610.13885
AB - Current AI predicts materials by speeding up energy minimisation and training on large databases. However, scaling data generation prevents manual analysis by domain experts. If novelty is evaluated by chemical compositions or traditional representations in Crystallographic Information Files, or standard structure matchers based on manually chosen thresholds, one can easily generate millions of ``new'' compositions or files without discovering truly novel materials. This paper uses a hierarchy of crystal invariants to continuously trace and quantify structural novelty based on the new mathematical definition of a crystal structure and asymptotic guarantees for invariant descriptors. The invariant-based analysis through cascade computations on a modest computer exposes the unexpected scale of duplication in the world's largest materials databases, shown here in the high-profile case of GNoME datasets. To prevent current ``embarrassment of over-prediction'', we propose necessary conditions for next-generation AI that should be automatically enforced before claiming any new materials.
ER -