MD17

MD17 is a benchmark dataset that contains molecular dynamics simulations used to evaluate the performance of machine learning models, particularly in predicting molecular forces and energies.

Papers using MD17 (31)

High-order Equivariant Flow Matching for Density Functional Theory Hamiltonian Prediction2025 · 6 citesGFFMERGE: Efficient Merging of Graph Neural Force Fields and Beyond2026Global properties of the energy landscape: a testing and training arena for machine learned potentials2025 · 2 citesPotential Score Matching: Debiasing Molecular Structure Sampling with Potential Energy Guidance2025 · 1 citesGradient-Guided Furthest Point Sampling for Robust Training Set Selection2025Improving Molecular Force Fields with Minimal Temporal Information2026Universal and efficient graph neural networks with dynamic attention for machine learning interatomic potentials2026Machine Learning Hamiltonians are Accurate Energy-Force Predictors2026ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamics20253D-GSRD: 3D Molecular Graph Auto-Encoder with Selective Re-mask Decoding2025Layer-to-Layer Knowledge Mixing in Graph Neural Network for Chemical Property Prediction2025Learning 3D Anisotropic Noise Distributions Improves Molecular Force Field Modeling2025DEQuify your force field: More efficient simulations using deep equilibrium models2025Efficient Prediction of SO(3)-Equivariant Hamiltonian Matrices via SO(2) Local Frames2025GeoRecon: Graph-Level Representation Learning for 3D Molecules via Reconstruction-Based Pretraining2025Beyond Force Metrics: Pre-Training MLFFs for Stable MD Simulations2025A Clifford Algebraic Approach to E(n)-Equivariant High-order Graph Neural Networks2024Learning Equivariant Non-Local Electron Density Functionals2024Ensemble Learning of Machine Learning Force Fields2024Accurate neural-network-based fitting of full-dimensional two-body potential energy surfaces2022ViSNet: an equivariant geometry-enhanced graph neural network with vector-scalar interactive message passing for molecules2022 · 3 citesEfficient and Equivariant Graph Networks for Predicting Quantum Hamiltonian2023 · 3 citesFractional Denoising for 3D Molecular Pre-training2023 · 3 citesQeMFi: A Multifidelity Dataset of Quantum Chemical Properties of Diverse Molecules2024 · 2 citesEnd-to-end AI framework for interpretable prediction of molecular and crystal properties2022 · 1 citesBeyond MD17: the reactive xxMD dataset2023 · 1 citesTransfer learning for chemically accurate interatomic neural network potentials2022No Headache for PIPs: A PIP Potential for Aspirin Outperforms Other Machine-Learned Potentials2024SE3Set: Harnessing equivariant hypergraph neural networks for molecular representation learning2024FreeCG: Free the Design Space of Clebsch-Gordan Transform for Machine Learning Force Fields2024Distribution Learning for Molecular Regression2024