MatMMExtract pipeline creates MatSciFig dataset of 391k annotated materials science figure panels and MaterialScope detection dataset with high accuracy.
Baseline reference
Jain , author S
Baseline reference. 50% of citing Pith papers use this work as a benchmark or comparison.
44
Pith papers citing it
11.6k
external citations · Crossref
Baseline
50% of classified citations
citation-role summary
dataset 7
background 3
citation-polarity summary
authors
co-cited works
representative citing papers
SLayerGen generates crystals invariant to any space or layer group via autoregressive lattice and Wyckoff sampling plus equivariant diffusion, achieving gains over bulk models on diperiodic materials after correcting a prior loss inconsistency for hexagonal groups.
A Gaussian process surrogate gate inserted between generative crystal models and property oracles matches or exceeds ungated fine-tuning while using roughly one-fifth the oracle calls for heat capacity and bulk modulus.
MP2SS reduces finite-size errors in periodic MP2 to millihartree accuracy at coarser k-point meshes for gapped systems via auxiliary function subtraction.
Constraint-aware neural networks clone known semilocal XC functionals more accurately in self-consistent calculations, transfer well from molecules to solids, and outperform unconstrained models across multiple tests.
LitXBench is a new benchmark for extracting complete experiments from scientific papers, with results showing frontier LLMs outperform multi-turn pipelines by up to 0.37 F1 due to better handling of processing steps.
AQVolt26 is a new high-temperature halide dataset that improves universal ML interatomic potentials for distorted configurations while showing that near-equilibrium relaxation data is not universally helpful.
Electrospinning-Data.org is a FAIR data platform that organizes electrospinning experiments into a structured, failure-inclusive corpus to enable predictive modeling and inverse design of nanofiber morphologies.
A new rapid synthesis method allows efficient production of isotope-enriched MoO3 crystals with control over both Mo and O isotopic content for phonon engineering.
Micromagnetic simulations of magnetoelastic coupling in exchange-decoupled Co/Ni islets predict a 52 dB/mm change in SAW transmission at 3.8 GHz depending on the magnetic state of neighboring islets.
IRNet uses per-layer residual shortcuts in fully connected networks to achieve better prediction accuracy and training convergence than prior ML methods on OQMD and Materials Project datasets for material properties.
Presents rules for constructing (NxH)-polytypes with La2WO6 stoichiometry and DFT results identifying La2WO6.oP36 as ground state with La2WO6.mC72 only 5 meV/at above hull.
A Stoner-inspired preconditioner based on non-interacting susceptibility that neglects orbital variations reduces SCF iterations in magnetic KS-DFT near phase transitions.
INCARBench evaluates 19 LLMs on VASP INCAR configuration generation and repair, showing high semantic accuracy but lower scientific correctness especially for DFT+U, magnetism, and correlated materials.
81-92% of chemically valid and metastable crystals from generative models are training duplicates or substitution-derived, with low-symmetry cases showing interpolation and high-symmetry cases showing memorization.
Pre-registered validation of an ML Na-cathode voltage screen yields 0.67 V MAE against experiment, with Materials Project PBE+U references 0.54 V low and dominating the error.
RADAR-PD introduces a modality-aware ML system that generates phase hypotheses from elemental constraints and performs recursive multiphase analysis with physics-constrained verification on experimental diffraction data.
Machine learning models, especially certain deep neural networks, can predict lattice thermal conductivity with useful accuracy across different generalization tests while being orders of magnitude faster than first-principles calculations.
A facet-resolved adsorption energy distribution method with ML force fields identifies active and methanol-selective alloy nanocatalyst surfaces for CO2 hydrogenation.
Multi-fidelity bandits screen 529 ZnO co-dopants with 81% fewer DFT calls, identify Y2Cu2 co-doped ZnO (1.84 eV) as optimal for visible-light band gaps, and release all 583 calculations plus code.
Physics-informed graph attention networks predict multi-phase equilibria in Ag-Bi-Cu-Sn alloys with 96% exact-set accuracy on in-domain data and strong generalization to unseen sections.
An ontology-aligned framework for atomistic simulations that integrates over 750,000 triples to enable interoperable data querying and automated provenance tracking.
deCIFer trains an autoregressive LM on 2.3 million structures with synthetic PXRD noise to generate CIF files, reporting 94% structural match rate on synthetic inorganic test sets.
Atompack delivers 96x faster shuffled reads and 79% smaller artifacts than ASE LMDB baselines for complete-record atomistic ML training workloads.
citing papers explorer
-
Equivariant Graph Neural Networks Improve Optical Spectra Prediction for Materials Screening
Equivariant GNNs outperform prior models on optical spectra and static permittivity prediction using RPA datasets for materials screening.