A.1 Graph-building strategies The graphs were built using the IsayevNN class from the pymatgen [48] package. It implements the commonly used Voronoi tessalation to define neighbors. Two atoms are considered bonded if they share a face in the Voronoi tessalation of the supercell and their distance is less than the sum of the atomic Cordero radii (a measure of the atomic radius) plus a cutoff =0 .5Å. This value of the cutoff was increase compared to [32] to reduce the number of disconnected graphs. We provide statistics for the graphs obtained by the method described in Section 5. A hard cutoff on atomic distances of 6Å is also imposed on atomic distances. Figure 5: Histogram of the number of primitive cell sites per material in the processed Materials Project dataset.

Supervised learning with deep models has tremendous potential for applications in materials science. Recently, graph neural networks have been used in this context, drawing direct inspiration from models for molecules. However, materials are typically much more structured than molecules, which is a feature that these models do not leverage. In this work, we introduce a class of models that are equivariant with respect to crystalline symmetry groups. We do this by defining a generalization of the message passing operations that can be used with more general permutation groups, or that can alternatively be seen as defining an expressive convolution operation on the crystal graph. Empirically, these models achieve competitive results with state-of-the-art on property prediction tasks.

Recent advances in generative machine learning have opened new possibilities for the discovery and design of novel materials. However, as these models become more sophisticated, the need for rigorous and meaningful evaluation metrics has grown. Existing evaluation approaches often fail to capture both the quality and novelty of generated structures, limiting our ability to assess true generative performance. In this paper, we introduce the Transport Novelty Distance (TNovD) to judge generative models used for materials discovery jointly by the quality and novelty of the generated materials. Based on ideas from Optimal Transport theory, TNovD uses a coupling between the features of the training and generated sets, which is refined into a quality and memorization regime by a threshold. The features are generated from crystal structures using a graph neural network that is trained to distinguish between materials, their augmented counterparts, and differently sized supercells using contrastive learning. We evaluate our proposed metric on typical toy experiments relevant for crystal structure prediction, including memorization, noise injection and lattice deformations. Additionally, we validate the TNovD on the MP20 validation set and the WBM substitution dataset, demonstrating that it is capable of detecting both memorized and low-quality material data. We also benchmark the performance of several popular material generative models. While introduced for materials, our TNovD framework is domain-agnostic and can be adapted for other areas, such as images and molecules.

Evaluating foundation models for crystallographic reasoning requires benchmarks that isolate generalization behavior while enforcing physical constraints. This work introduces a multiscale multicrystal dataset with two physically grounded evaluation protocols to stress-test multimodal generative models. The Spatial-Exclusion benchmark withholds all supercells of a given radius from a diverse dataset, enabling controlled assessments of spatial interpolation and extrapolation. The Compositional-Exclusion benchmark omits all samples of a specific chemical composition, probing generalization across stoichiometries. Nine vision--language foundation models are prompted with crystallographic images and textual context to generate structural annotations. Responses are evaluated via (i) relative errors in lattice parameters and density, (ii) a physics-consistency index penalizing volumetric violations, and (iii) a hallucination score capturing geometric outliers and invalid space-group predictions. These benchmarks establish a reproducible, physically informed framework for assessing generalization, consistency, and reliability in large-scale multimodal models. Dataset and code are available at https://github.com/KurbanIntelligenceLab/StressTestingMMFMinCR.

The attention mechanism was originally introduced in the context of large language models to learn relations between words [26]. Solving the many-electron Schrödinger equation for Here, the attention mechanism is employed to identify solids is an exceedingly difficult problem due to the exponential and quantify how electrons influence each other and how growth of the Hilbert space dimension. Various such influence affects their individual orbitals. This enable techniques based on the variational principle have long the construction of NN wavefunctions from Slater been developed to approximate the ground state of interacting determinants of generalized orbitals that depend on the electron systems using trial wavefunctions.

Phonon-assisted optical absorption in semiconductors is crucial for understanding and optimizing optoelectronic devices, yet its accurate simulation remains a significant challenge in computational materials science. We present an efficient approach that combines deep learning tight-binding (TB) and potential models to efficiently calculate the phonon-assisted optical absorption in semiconductors with $ab$ $initio$ accuracy. Our strategy enables efficient sampling of atomic configurations through molecular dynamics and rapid computation of electronic structure and optical properties from the TB models. We demonstrate its efficacy by calculating the temperature-dependent optical absorption spectra and band gap renormalization of Si and GaAs due to electron-phonon coupling over a temperature range of 100-400 K. Our results show excellent agreement with experimental data, capturing both indirect and direct absorption processes, including subtle features like the Urbach tail. This approach offers a powerful tool for studying complex materials with high accuracy and efficiency, paving the way for high-throughput screening of optoelectronic materials.

We present a machine learning (ML) method for efficient computation of vibrational thermal expectation values of physical properties from first principles. Our approach is based on the non-perturbative frozen phonon formulation in which stochastic Monte Carlo algorithm is employed to sample configurations of nuclei in a supercell at finite temperatures based on a first-principles phonon model. A deep-learning neural network is trained to accurately predict physical properties associated with sampled phonon configurations, thus bypassing the time-consuming {\em ab initio} calculations. To incorporate the point-group symmetry of the electronic system into the ML model, group-theoretical methods are used to develop a symmetry-invariant descriptor for phonon configurations in the supercell. We apply our ML approach to compute the temperature dependent electronic energy gap of silicon based on density functional theory (DFT). We show that, with less than a hundred DFT calculations for training the neural network model, an order of magnitude larger number of sampling can be achieved for the computation of the vibrational thermal expectation values. Our work highlights the promising potential of ML techniques for finite temperature first-principles electronic structure methods.

Deep-Learning-based Variational Monte Carlo (DL-VMC) has recently emerged as a highly accurate approach for finding approximate solutions to the many-electron Schr\"odinger equation. Despite its favorable scaling with the number of electrons, $\mathcal{O}(n_\text{el}^{4})$, the practical value of DL-VMC is limited by the high cost of optimizing the neural network weights for every system studied. To mitigate this problem, recent research has proposed optimizing a single neural network across multiple systems, reducing the cost per system. Here we extend this approach to solids, where similar but distinct calculations using different geometries, boundary conditions, and supercell sizes are often required. We show how to optimize a single ansatz across all of these variations, reducing the required number of optimization steps by an order of magnitude. Furthermore, we exploit the transfer capabilities of a pre-trained network. We successfully transfer a network, pre-trained on 2x2x2 supercells of LiH, to 3x3x3 supercells. This reduces the number of optimization steps required to simulate the large system by a factor of 50 compared to previous work.

The first-principles-based effective Hamiltonian is widely used to predict and simulate the properties of ferroelectrics and relaxor ferroelectrics. However, the parametrization method of the effective Hamiltonian is complicated and hardly can resolve the systems with complex interactions and/or complex components. Here, we developed an on-the-fly machine learning approach to parametrize the effective Hamiltonian based on Bayesian linear regression. The parametrization is completed in molecular dynamics simulations, with the energy, forces and stress predicted at each step along with their uncertainties. First-principles calculations are executed when the uncertainties are large to retrain the parameters. This approach provides a universal and automatic way to compute the effective Hamiltonian parameters for any considered systems including complex systems which previous methods can not handle. BaTiO3 and Pb(Sc,Ta)O3 are taken as examples to show the accurateness of this approach comparing with conventional first-principles parametrization method.