arXiv subjects

Stefano Falletta

Publications and source records attributed to Stefano Falletta.

7 recordsLinked to original sources

arXiv2026

CrystalReasoner: Reasoning and RL for Property-Conditioned Crystal Structure Generation

Generative modeling has emerged as a promising approach for crystal structure discovery. However, existing LLM-based generative models struggle with low-level atomic precision, while diffusion-based methods fall short in integrating high-level scientific knowledge. As a result, generated structures are often invalid, unstable, or do not possess desirable properties. To address this gap, we propose CrystalReasoner (CrysReas), an end-to-end LLM framework that generates crystal structures from natural language instructions through reasoning and alignment. CrysReas introduces physical priors as thinking tokens, which include crystallographic symmetry, local coordination environments and predicted physical properties before generating atomic coordinates. This bridges the gap between natural language and 3D structures. CrysReas then employs reinforcement learning (RL) with a multi-objective, dense reward function to align generation with physical validity, chemical consistency, and thermodynamic stability. For property-conditioned tasks, we design task-specific reward functions and train specialized models for discrete constraints (e.g., space group) and continuous properties (e.g., elasticity, thermal expansion). Empirical results demonstrate that compared to prior works and baselines without thinking traces or RL, CrysReas obtains better performance on diverse metrics, triples S.U.N. ratio, and achieves better performance for property conditioned generation. CrysReas also exhibits adaptive reasoning, increasing reasoning lengths as the number of atoms increases. Our work demonstrates the potential of leveraging thinking traces and RL for generating valid, stable, and property-conditioned crystal structures.

cs.AI↗

arXiv2025

Equivalence of charged and neutral density functional formulations for correcting the many-body self-interaction of polarons

The electron self-interaction problem in density functional theory affects the accurate modeling of polarons, particularly their localization and formation energy. Charged and neutral density functional formulations have been developed to address this issue, yet their relationship remains unclear. Here, we demonstrate their equivalence in treating the many-body self-interaction of the polaron state. In particular, we connect with each other piecewise-linear functionals based on adding an extra charge to the supercell, the pSIC approach derived from the energetics of the neutral defect with polaronic distortions in a supercell, and the unit-cell method for polarons based on electron-phonon couplings. We show that these approaches lead to the same formal expression of the self-interaction corrected energy, which is fully defined by the energetics of the neutral charge state of the charged polaronic structure. Residual differences between these methods solely arise from the achieved polaronic structure, which is affected by different treatments of electron-screening and finite-size effects. We apply these methods to a set of prototypical small hole and electron polarons, including the hole polaron in MgO, the hole polaron in $β$-Ga$_2$O$_3$, the $V_\text{k}$ center in NaI, the electron polaron in BiVO$_4$, and the electron polaron in TiO$_2$. We show that the ground-state properties of polarons obtained using charged and neutral density functional formulations are in excellent agreement.

cond-mat.mtrl-sci↗

arXiv2025

TorchSim: An efficient atomistic simulation engine in PyTorch

We introduce TorchSim, an open-source atomistic simulation engine tailored for the Machine Learned Interatomic Potential (MLIP) era. By rewriting core atomistic simulation primitives in PyTorch, TorchSim can achieve orders of magnitude acceleration for popular MLIPs. Unlike existing molecular dynamics packages, which simulate one system at a time, TorchSim performs batched simulations that efficiently utilize modern GPUs by evolving multiple systems concurrently. TorchSim supports molecular dynamics integrators, structural relaxation optimizers, both machine-learned and classical interatomic potentials (such as Lennard-Jones, Morse, soft-sphere), batching with automatic memory management, differentiable simulation, and integration with popular materials informatics tools.

physics.comp-ph↗

arXiv2025

Room-temperature decomposition of the ethaline deep eutectic solvent

Environmentally-benign, non-toxic electrolytes with combinatorial design spaces are excellent candidates for green solvents, green leaching agents, and carbon capture sources. Here, we examine one particular green solvent, ethaline, a 2:1 molar ratio of ethylene glycol and choline chloride. Despite its touted green credentials, we find partial decomposition of ethaline into toxic chloromethane and dimethylaminoethanol at room temperature, limiting its sustainable advantage. We experimentally characterize these decomposition products and computationally develop a general, quantum chemically-accurate workflow to understand decomposition. We find that fluctuations of the hydrogen bonds bind chloride near reaction sites, initiating the reaction between choline cations and chloride anions. In summary, in the design of green solvents, we do not recommend the use of choline chloride due to its susceptibility to undergo decomposition in strongly hydrogen-bound mixtures.

physics.chem-ph↗

arXiv2024

Unified Differentiable Learning of Electric Response

Predicting response of materials to external stimuli is a primary objective of computational materials science. However, current methods are limited to small-scale simulations due to the unfavorable scaling of computational costs. Here, we implement an equivariant machine-learning framework where response properties stem from exact differential relationships between a generalized potential function and applied external fields. Focusing on responses to electric fields, the method predicts electric enthalpy, forces, polarization, Born charges, and polarizability within a unified model enforcing the full set of exact physical constraints, symmetries and conservation laws. Through application to $α$-SiO$_2$, we demonstrate that our approach can be used for predicting vibrational and dielectric properties of materials, and for conducting large-scale dynamics under arbitrary electric fields at unprecedented accuracy and scale. We apply our method to ferroelectric BaTiO$_3$ and capture the temperature-dependence and time evolution of hysteresis, revealing the underlying microscopic mechanisms of nucleation and growth that govern ferroelectric domain switching.

cond-mat.mtrl-sci↗

arXiv2024

Addressing the Band Gap Problem with a Machine-Learned Exchange Functional

The systematic underestimation of band gaps is one of the most fundamental challenges in semilocal density functional theory (DFT). In addition to hindering the application of DFT to predicting electronic properties, the band gap problem is intimately related to self-interaction and delocalization errors, which make the study of charge transfer mechanisms with DFT difficult. In this work, we present two key innovations to address the band gap problem. First, we design an approach for machine learning density functionals based on Gaussian processes to explicitly fit single-particle energy levels. Second, we introduce novel nonlocal features of the density matrix that are expressive enough to fit these single-particle levels. Combining these developments, we train a machine-learned functional for the exact exchange energy that predicts molecular energy gaps and reaction energies of a wide range of molecules in excellent agreement with reference hybrid DFT calculations. In addition, while being trained solely on molecular data, our model predicts reasonable formation energies of polarons in solids, showcasing its transferability and robustness. Our approach generalizes straightforwardly to full exchange-correlation functionals, thus paving the way to the design of novel state-of-the-art functionals for the prediction of electronic properties of molecules and materials.

physics.chem-ph↗

arXiv2022

Hubbard $U$ through polaronic defect states

Since the preliminary work of Anisimov and co-workers, the Hubbard corrected DFT+$U$ functional has been used for predicting properties of correlated materials by applying on-site effective Coulomb interactions to specific orbitals. However, the determination of the Hubbard $U$ parameter has remained under intense discussion despite the multitude of approaches proposed. Here, we define a selection criterion based on the use of polaronic defect states for the enforcement of the piecewise linearity of the total energy upon electron occupation. A good agreement with results from piecewise linear hybrid functionals is found for the electronic and structural properties of polarons, including the formation energies. The values of $U$ determined in this way are found to give a robust description of the polaron energetics upon variation of the considered state. In particular, we also address a polaron hopping pathway, finding that the determined value of $U$ leads to accurate energetics without requiring a configurational-dependent $U$. It is emphasized that the selection of $U$ should be based on physical properties directly associated with the orbitals to which $U$ is applied, rather than on more global properties such as band gaps and band widths. For comparison, we also determine $U$ through a well-established linear-response scheme finding noticeably different values of $U$ and consequently different formation energies. Possible origins of these discrepancies are discussed. As case studies, we consider the self-trapped electron in BiVO$_4$, the self-trapped hole in MgO, the Li-trapped hole in MgO, and the Al-trapped hole in $α$-SiO$_2$.

cond-mat.mtrl-sci↗