ArBG replaces flow-based methods with autoregressive models for Boltzmann sampling, showing gains on peptide benchmarks and a 132M-parameter model Robin cutting zero-shot energy error by over 60% on 8-residue systems.
hub
The atomic simulation environment—a Python library for working with atoms.Journal of Physics: Condensed Mat- ter, 29(27):273002, June 2017
17 Pith papers cite this work, alongside 3,266 external citations. Polarity classification is still indexing.
17
Pith papers citing it
3,266
external citations · Crossref
hub tools
citation-role summary
method 1
citation-polarity summary
roles
method 1polarities
use method 1representative citing papers
Ion irradiation creates a layer-disordered phase with planar defects in MnBi2Te4 that suppresses symmetry and anomalous Hall conductivity fivefold while magnetic order persists.
MLM is an open-source Python package that uses a solve-and-round algorithm to construct commensurate moire supercells for arbitrary multilayer twisted 2D systems with O(N^2) scaling per twist angle.
A physics-plus-AI pipeline recovers crystal structures from powder X-ray diffraction on most of a hard benchmark set, but success drops sharply for low-symmetry cases.
Drift-React produces full minimum energy pathways for reactions in a single step via SE(3) drifting fields, matching TS accuracy of iterative models with orders-of-magnitude speedup on Transition1x and Halo8 datasets.
A physics-informed autoencoder compresses 3D charge density into a 16x16x16x16 latent representation that, combined with MAGPIE descriptors, predicts bulk modulus, Young's modulus, shear modulus, formation energy, and Debye temperature with R2 values of 0.94, 0.88, 0.87, 0.96, and 0.89 on 6059 DFT-s
QCOF ML potentials tuned on COF data outperform general MACE models for defective systems and reveal higher thermal defect sensitivity in CTF-1 versus COF-LZU1 with nearly invariant low-strain mechanics.
An ontology-aligned framework for atomistic simulations that integrates over 750,000 triples to enable interoperable data querying and automated provenance tracking.
QUASAR is a new autonomous LLM-based system that orchestrates multi-scale atomistic simulations and benchmarks as a general reasoning tool rather than a narrow automation script.
Atompack delivers 96x faster shuffled reads and 79% smaller artifacts than ASE LMDB baselines for complete-record atomistic ML training workloads.
Hybrid QM/ML forcefield framework couples DFT with MLIPs to enable scalable, chemically accurate simulations of solute-dislocation interactions, demonstrated on Sn/Fe segregation in Zr and magnetic effects in steel.
Transition path sampling serves as an active learning engine to build machine-learned potentials accurate in barrier regions, enabling discovery of multiple protonation mechanisms in CO2 reduction on copper.
DFT and cluster-expansion calculations identify 4.21 Å as the threshold above which Na intercalates readily in graphite-like carbon while Li capacity peaks narrowly near 3.75 Å with AA stacking preferred for both ions.
Machine-learned force fields trained on coupled-cluster potential energy surfaces produce phonon dispersions and vibrational densities of states for solids that agree better with experiment than DFT-based models.
A pipeline samples site-disordered material configurations with 400 virtual cells when the supercell is large enough, improving computational feasibility over quasirandom or cluster expansion methods.
DFT calculations find Si diffusion barriers in GaN of 3.2 eV to 9.9 eV by direction with experiments confirming negligible diffusion even at 1450°C and 1 GPa.
Bayesian optimization with Gaussian processes unifies minimization, single-point saddle searches, and double-ended path searches on potential energy surfaces through a shared six-step surrogate loop using derivative observations and inverse-distance kernels.
citing papers explorer
-
Autoregressive Boltzmann Generators
ArBG replaces flow-based methods with autoregressive models for Boltzmann sampling, showing gains on peptide benchmarks and a 132M-parameter model Robin cutting zero-shot energy error by over 60% on 8-residue systems.
-
Disorder-driven symmetry suppression by van der Waals planar defects in a magnetic topological insulator
Ion irradiation creates a layer-disordered phase with planar defects in MnBi2Te4 that suppresses symmetry and anomalous Hall conductivity fivefold while magnetic order persists.
-
MLM: Multi-Layer Moire -- A Python Package for Generating Commensurate Supercells of Twisted Multilayer Two-Dimensional Materials
MLM is an open-source Python package that uses a solve-and-round algorithm to construct commensurate moire supercells for arbitrary multilayer twisted 2D systems with O(N^2) scaling per twist angle.
-
Ab-initio Crystal Structure Determination from Powder X-Ray Diffraction
A physics-plus-AI pipeline recovers crystal structures from powder X-ray diffraction on most of a hard benchmark set, but success drops sharply for low-symmetry cases.
-
Drift-React: One-step Generation of Reaction Pathways via SE(3) Drifting Fields
Drift-React produces full minimum energy pathways for reactions in a single step via SE(3) drifting fields, matching TS accuracy of iterative models with orders-of-magnitude speedup on Transition1x and Halo8 datasets.
-
Physics Aware Representation Learning on Electronic Charge Density for Materials Property Prediction
A physics-informed autoencoder compresses 3D charge density into a 16x16x16x16 latent representation that, combined with MAGPIE descriptors, predicts bulk modulus, Young's modulus, shear modulus, formation energy, and Debye temperature with R2 values of 0.94, 0.88, 0.87, 0.96, and 0.89 on 6059 DFT-s
-
Data-Driven Thermal and Mechanical Modeling of Defective Covalent Organic Frameworks
QCOF ML potentials tuned on COF data outperform general MACE models for defective systems and reveal higher thermal defect sensitivity in CTF-1 versus COF-LZU1 with nearly invariant low-strain mechanics.
-
Ontology-based knowledge graph infrastructure for interoperable atomistic simulation data
An ontology-aligned framework for atomistic simulations that integrates over 750,000 triples to enable interoperable data querying and automated provenance tracking.
-
QUASAR: A Universal Autonomous System for Atomistic Simulation and a Benchmark of Its Capabilities
QUASAR is a new autonomous LLM-based system that orchestrates multi-scale atomistic simulations and benchmarks as a general reasoning tool rather than a narrow automation script.
-
Atompack: A Storage and Distribution Layer for Read-Heavy Atomistic ML Training Datasets
Atompack delivers 96x faster shuffled reads and 79% smaller artifacts than ASE LMDB baselines for complete-record atomistic ML training workloads.
-
A Hybrid Quantum Mechanics Machine Learning Forcefield (QM/ML) Framework for Accurate Solute-Dislocation Interaction Simulations
Hybrid QM/ML forcefield framework couples DFT with MLIPs to enable scalable, chemically accurate simulations of solute-dislocation interactions, demonstrated on Sn/Fe segregation in Zr and magnetic effects in steel.
-
Discovering Reaction Mechanisms with Transition Path Sampling-Based Active Learning of Machine-Learned Potentials
Transition path sampling serves as an active learning engine to build machine-learned potentials accurate in barrier regions, enabling discovery of multiple protonation mechanisms in CO2 reduction on copper.
-
Rational Design Principles for Na- and Li-ion Carbon Anodes from Interlayer Spacing Control
DFT and cluster-expansion calculations identify 4.21 Å as the threshold above which Na intercalates readily in graphite-like carbon while Li capacity peaks narrowly near 3.75 Å with AA stacking preferred for both ions.
-
Machine-Learned Force Fields for Lattice Dynamics at Coupled-Cluster Level Accuracy
Machine-learned force fields trained on coupled-cluster potential energy surfaces produce phonon dispersions and vibrational densities of states for solids that agree better with experiment than DFT-based models.
-
Virp: neural network-accelerated prediction of physical properties in site-disordered materials
A pipeline samples site-disordered material configurations with 400 virtual cells when the supercell is large enough, improving computational feasibility over quasirandom or cluster expansion methods.
-
Limited Diffusion of Silicon in GaN: A DFT Study Supported by Experimental Evidence
DFT calculations find Si diffusion barriers in GaN of 3.2 eV to 9.9 eV by direction with experiments confirming negligible diffusion even at 1450°C and 1 GPa.
-
A Tutorial Review of Bayesian Optimization with Gaussian Processes to Accelerate Stationary Point Searches
Bayesian optimization with Gaussian processes unifies minimization, single-point saddle searches, and double-ended path searches on potential energy surfaces through a shared six-step surrogate loop using derivative observations and inverse-distance kernels.