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.
Dotson, Raimondas Galvelis, John E
8 Pith papers cite this work, alongside 168 external citations. Polarity classification is still indexing.
8
Pith papers citing it
168
external citations · OpenAlex
years
2026 8verdicts
UNVERDICTED 8representative citing papers
SOAP and SOAP-Muon optimizers deliver faster convergence and higher final accuracy than Adam for NequIP and Allegro MLIPs, with the largest gains under partial force supervision.
SKMD adapts Stein variational gradient descent into molecular dynamics with asynchronous updates and global atomic descriptor kernels to acquire non-redundant training configurations while preserving the Boltzmann distribution, yielding higher MLIP accuracy with fewer samples than baselines.
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.
THEMol is a new large dataset of torsion, Hessian, energy, and multipole data from DFT for closed-shell organic molecules, organized into five subsets for use in molecular potential development.
ORION is a universal ML force field for organic molecules that reaches near-DFT accuracy on forces while running 215 times faster than ReaxFF by using a unified top-down and bottom-up training dataset strategy.
Benchmarks of 15 MLIPs show parameter count and training set size correlate with accuracy, architecture drives speed and memory, and explicit Coulomb terms provide no benefit.
mlip v2 is a new software release that integrates API redesign, e3j backend, eSEN model, improved charge modeling, and expanded simulation capabilities to support larger-scale molecular modeling.
citing papers explorer
-
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.
-
Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials
SOAP and SOAP-Muon optimizers deliver faster convergence and higher final accuracy than Adam for NequIP and Allegro MLIPs, with the largest gains under partial force supervision.
-
Stein Kernelized Molecular Dynamics for Active Learning of Interatomic Potentials
SKMD adapts Stein variational gradient descent into molecular dynamics with asynchronous updates and global atomic descriptor kernels to acquire non-redundant training configurations while preserving the Boltzmann distribution, yielding higher MLIP accuracy with fewer samples than baselines.
-
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.
-
THEMol dataset: Torsion, Hessian, and Energy of Molecules
THEMol is a new large dataset of torsion, Hessian, energy, and multipole data from DFT for closed-shell organic molecules, organized into five subsets for use in molecular potential development.
-
ORION: Unifying Top-Down and Bottom-Up Chemical Space Sampling for a Universal Organic Force Field
ORION is a universal ML force field for organic molecules that reaches near-DFT accuracy on forces while running 215 times faster than ReaxFF by using a unified top-down and bottom-up training dataset strategy.
-
Accuracy and Efficiency Benchmarks of Pretrained Machine Learning Potentials for Molecular Simulations
Benchmarks of 15 MLIPs show parameter count and training set size correlate with accuracy, architecture drives speed and memory, and explicit Coulomb terms provide no benefit.
-
Machine Learning Interatomic Potentials: Advancing Open-Source Software for Efficient and Scalable Molecular Simulation
mlip v2 is a new software release that integrates API redesign, e3j backend, eSEN model, improved charge modeling, and expanded simulation capabilities to support larger-scale molecular modeling.