Probing the limits of forward screening and inverse design of crystal materials
Abstract
Crystal materials are designed either by screening known crystals for a target property or by generating new crystals with that property. Both routes are judged against public databases of known crystals, but these databases name the same crystal differently. Several materials announced as new by recent generative and autonomous discovery campaigns have since proved to be known phases or close variants of them. The limits of both routes remain unmeasured. Here we probe them by matching the crystals of 55 databases by structure into one self-extending database of 14.6 million crystals and 131.8 million property values, each traced to its source. Only 4.5% of its crystal-property pairs carry a value. To fill the empty pairs, we developed CGTNet, an equivariant graph transformer with state-of-the-art accuracy on several tasks of two public benchmarks. Trained on each task, it forms a library of property models whose predictions we release. The reach of forward screening has been underestimated. Known crystals hold tens of thousands of ultra-incompressible candidates without a recorded bulk modulus, and a post hoc audit of synthesis evidence finds among them ReN2, whose bulk modulus was measured at 428 GPa. The novelty claimed for inverse design has been overestimated. At an equal verification budget, a stiffness-conditioned generator returned more stiff hits than screening (100 versus 55), but all contained osmium and 82 repeated recorded compositions. Integrating and completing the databases of known crystals therefore moves the limits of both routes: it widens what screening can reach and sets the reference against which generated crystals are judged new.
Related
- Symmetry- and Property-Aware Crystal Generation with Reinforcement Learning for Inverse Materials Design
- ALIGNN 2.0: A Unified Line-Graph Neural Network Framework for Materials Screening, Force Fields, Inverse Design, Spectroscopy, and Microscopy
- Probing Saturation Effect in Heavy Meson Pair Correlation in Forward pA Collisions
- Physics-Augmented Graph Transformers for Patch-Antenna Forward and Inverse Design
- Probing Forward-Backward Multiplicity Correlations and Fluctuations Using the Strongly Intensive Observable Σ in Pb-Pb Collisions at SPS Energies with UrQMD
BibTeX
@misc{song2026probing,
title = {{Probing the limits of forward screening and inverse design of crystal materials}},
author = {Zhilong Song and Zijun Pang and Lixue Cheng},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.32678},
eprint = {2610.32678},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.32678},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Probing the limits of forward screening and inverse design of crystal materials
AU - Zhilong Song
AU - Zijun Pang
AU - Lixue Cheng
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.32678
UR - https://arxiv.org/abs/2610.32678
AB - Crystal materials are designed either by screening known crystals for a target property or by generating new crystals with that property. Both routes are judged against public databases of known crystals, but these databases name the same crystal differently. Several materials announced as new by recent generative and autonomous discovery campaigns have since proved to be known phases or close variants of them. The limits of both routes remain unmeasured. Here we probe them by matching the crystals of 55 databases by structure into one self-extending database of 14.6 million crystals and 131.8 million property values, each traced to its source. Only 4.5% of its crystal-property pairs carry a value. To fill the empty pairs, we developed CGTNet, an equivariant graph transformer with state-of-the-art accuracy on several tasks of two public benchmarks. Trained on each task, it forms a library of property models whose predictions we release. The reach of forward screening has been underestimated. Known crystals hold tens of thousands of ultra-incompressible candidates without a recorded bulk modulus, and a post hoc audit of synthesis evidence finds among them ReN2, whose bulk modulus was measured at 428 GPa. The novelty claimed for inverse design has been overestimated. At an equal verification budget, a stiffness-conditioned generator returned more stiff hits than screening (100 versus 55), but all contained osmium and 82 repeated recorded compositions. Integrating and completing the databases of known crystals therefore moves the limits of both routes: it widens what screening can reach and sets the reference against which generated crystals are judged new.
ER -