A differentiable pipeline uses continuous atom occupancy and gradient descent plus a neural network to optimize short-range order in multi-element alloys directly for target stiffness properties.
hub Canonical reference
Nature Machine Intelligence5(9), 1031– 1041 (2023)
Canonical reference. 71% of citing Pith papers cite this work as background.
14
Pith papers citing it
725
external citations · Crossref
Background
71% of classified citations
hub tools
citation-role summary
background 5
baseline 2
citation-polarity summary
years
2026 14representative citing papers
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.
Kernels from pretrained MLIP latent spaces outperform standard acquisition methods in active learning for reactive chemistry, reducing required labels by 38% for energy error and 28% for force error.
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.
Interfacial energetics computed with machine-learned potentials explain why Mn addition switches grain-boundary precipitates in Cu-Ni-Si alloys from irregular Ni2Si to film-shaped Mn6Ni16Si7.
Force-aware Neural Tangent Kernels combined with chunked acquisition provide scalable and distribution-robust active learning for MLIPs, outperforming baselines on OC20 and remaining competitive on other benchmarks.
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.
Atompack delivers 96x faster shuffled reads and 79% smaller artifacts than ASE LMDB baselines for complete-record atomistic ML training workloads.
A pretrained universal MLIP without spin or Hubbard-U corrections predicts the same chemisorbed S removal and O uptake under 15 eV O+/O2+ bombardment of WS2 as after three rounds of PBE+D3+U+spin fine-tuning, which still reduces energy/force MAE to 4.5 meV/atom and 0.076 eV/A.
Synthetic pre-training on ML-generated tensor data followed by fine-tuning on ground-truth calculations improves data efficiency for graph models of solid-state NMR parameters when the pre-training and fine-tuning domains match.
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
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.
Implements thermodynamic models for pure elements from 0 K in PyCalphad and ESPEI, remodeling 41 elements with MCMC uncertainty quantification to support improved CALPHAD descriptions.
citing papers explorer
-
Differentiable inverse design of short-range order in high-entropy alloys: from target sro to target property
A differentiable pipeline uses continuous atom occupancy and gradient descent plus a neural network to optimize short-range order in multi-element alloys directly for target stiffness properties.
-
Surrogate-Gated Generation and Foundation-Model Embeddings for Bayesian Materials Design
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.
-
Pretrained Model Representations as Acquisition Signals for Active Learning of MLIPs
Kernels from pretrained MLIP latent spaces outperform standard acquisition methods in active learning for reactive chemistry, reducing required labels by 38% for energy error and 28% for force error.
-
AQVolt26: High-Temperature r$^2$SCAN Halide Dataset for Universal ML Potentials and Solid-State Batteries
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.
-
Precipitate phase selection and grain boundary morphology in Cu-Ni-Si-Mn alloys: A machine-learning interatomic potential study
Interfacial energetics computed with machine-learned potentials explain why Mn addition switches grain-boundary precipitates in Cu-Ni-Si alloys from irregular Ni2Si to film-shaped Mn6Ni16Si7.
-
Force-Aware Neural Tangent Kernels for Scalable and Robust Active Learning of MLIPs
Force-aware Neural Tangent Kernels combined with chunked acquisition provide scalable and distribution-robust active learning for MLIPs, outperforming baselines on OC20 and remaining competitive on other benchmarks.
-
Accelerated Dopant Screening in Oxide Semiconductors via Multi-Fidelity Contextual Bandits and a Three-Tier DFT Validation Funnel
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.
-
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.
-
Fine-Tuning a Universal Machine-Learned Interatomic Potential for Oxygen Plasma Interactions with WS$_2$
A pretrained universal MLIP without spin or Hubbard-U corrections predicts the same chemisorbed S removal and O uptake under 15 eV O+/O2+ bombardment of WS2 as after three rounds of PBE+D3+U+spin fine-tuning, which still reduces energy/force MAE to 4.5 meV/atom and 0.076 eV/A.
-
Synthetic pre-training of graph-network models for predicting solid-state NMR parameters
Synthetic pre-training on ML-generated tensor data followed by fine-tuning on ground-truth calculations improves data efficiency for graph models of solid-state NMR parameters when the pre-training and fine-tuning domains match.
-
Heterogeneous Scientific Foundation Model Collaboration
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
-
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.
-
Thermodynamic Modeling of Pure Elements from 0 K with Uncertainty Quantification using PyCalphad and ESPEI
Implements thermodynamic models for pure elements from 0 K in PyCalphad and ESPEI, remodeling 41 elements with MCMC uncertainty quantification to support improved CALPHAD descriptions.
- Spatial statistics for screening molecular structures