ALIGNN 2.0: A Unified Line-Graph Neural Network Framework for Materials Screening, Force Fields, Inverse Design, Spectroscopy, and Microscopy
Abstract
Graph neural networks are central to materials property prediction and machine-learning interatomic potentials, yet their reliance on specialized graph libraries hampers portability and reproducibility, and property and force-field models have historically required separate graph pipelines. We present ALIGNN 2.0, a dependency-free, pure-PyTorch reimplementation of the Atomistic Line Graph Neural Network, with the line graph and its batching built from scratch, running on current-generation accelerators and unifying scalar, spectral, tensorial, per-atom, and force-field prediction behind a single graph, a combination that to our knowledge no existing framework provides. Comparing radius and k-nearest-neighbor (kNN) graphs, the wider kNN graph is more accurate for properties while the smoothly varying radius graph is required for energy-conserving molecular dynamics. On the JARVIS-Leaderboard, ALIGNN 2.0 leads on 26 of 30 single-property benchmarks against the original ALIGNN, with large gains for piezoelectric and dielectric maxima, exfoliation energy, moduli, and superconducting Tc. The LAMMPS- and OpenMM-compatible ALIGNN-FF matches leading universal potentials on the Matbench-Discovery and CHIPS-FF benchmarks at a small fraction of their parameters while scaling to hundred-thousand-atom cells. We further use ALIGNN 2.0 as the denoiser in a conditional crystal-diffusion model, where explicit line-graph message passing consistently lowers structural denoising loss. We also show, as work in progress, that an independently diffused, redundant bond-angle state is learnable but does not uniformly improve reconstruction or combine additively with the line graph. Finally, from a single relaxed structure the same framework reconstructs infrared, Raman, optical-dielectric, and neutron spectra in agreement with experiment and DFT, and drives frozen-phonon electron-microscopy image simulation.
Related
- Development of a Physics-Informed Neural Framework, MEOWN, for Rapid Prediction of Muon Stopping Sites in Crystalline Materials, for understanding Quantum Magnet employing Muon Spectroscopy
- Modeling coherent nonlinear microscopy of axially layered anisotropic materials using FDTD
- Quantifying Phase Transformations in Alloying Anodes via in-situ Liquid Cell Hard X-ray Spectroscopy and Cryogenic Microscopy
- Ultrafast Electron Microscopy: A Quantitative Platform for Nonequilibrium Materials Research
- Multi-techniques characterisation of anti-reflective Ta2O5 and TaOxNy thin films deposited by reactive sputtering: coupling X-ray photoelectron spectroscopy, scanning/transmission electron microscopy and ion beam analysis
BibTeX
@misc{lee2026alignn,
title = {{ALIGNN 2.0: A Unified Line-Graph Neural Network Framework for Materials Screening, Force Fields, Inverse Design, Spectroscopy, and Microscopy}},
author = {Jaehyung Lee and Charles Rhys Campbell and Akshaya Ajith and Sergei V. Kalinin and Christopher Wolverton and Kamal Choudhary},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.19487},
eprint = {2609.19487},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.19487},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - ALIGNN 2.0: A Unified Line-Graph Neural Network Framework for Materials Screening, Force Fields, Inverse Design, Spectroscopy, and Microscopy
AU - Jaehyung Lee
AU - Charles Rhys Campbell
AU - Akshaya Ajith
AU - Sergei V. Kalinin
AU - Christopher Wolverton
AU - Kamal Choudhary
PY - 2026
DA - 2026/09/16
PB - arXiv
DO - 10.48550/arxiv.2609.19487
UR - https://arxiv.org/abs/2609.19487
AB - Graph neural networks are central to materials property prediction and machine-learning interatomic potentials, yet their reliance on specialized graph libraries hampers portability and reproducibility, and property and force-field models have historically required separate graph pipelines. We present ALIGNN 2.0, a dependency-free, pure-PyTorch reimplementation of the Atomistic Line Graph Neural Network, with the line graph and its batching built from scratch, running on current-generation accelerators and unifying scalar, spectral, tensorial, per-atom, and force-field prediction behind a single graph, a combination that to our knowledge no existing framework provides. Comparing radius and k-nearest-neighbor (kNN) graphs, the wider kNN graph is more accurate for properties while the smoothly varying radius graph is required for energy-conserving molecular dynamics. On the JARVIS-Leaderboard, ALIGNN 2.0 leads on 26 of 30 single-property benchmarks against the original ALIGNN, with large gains for piezoelectric and dielectric maxima, exfoliation energy, moduli, and superconducting Tc. The LAMMPS- and OpenMM-compatible ALIGNN-FF matches leading universal potentials on the Matbench-Discovery and CHIPS-FF benchmarks at a small fraction of their parameters while scaling to hundred-thousand-atom cells. We further use ALIGNN 2.0 as the denoiser in a conditional crystal-diffusion model, where explicit line-graph message passing consistently lowers structural denoising loss. We also show, as work in progress, that an independently diffused, redundant bond-angle state is learnable but does not uniformly improve reconstruction or combine additively with the line graph. Finally, from a single relaxed structure the same framework reconstructs infrared, Raman, optical-dielectric, and neutron spectra in agreement with experiment and DFT, and drives frozen-phonon electron-microscopy image simulation.
ER -