mlsanityData to 7 Oct 2026

Ewald-based Long-Range Message Passing for Molecular Graphs

arXiv 2303.04791 ↗First posted 8 Mar 2023cs.LG cond-mat.mtrl-sci physics.chem-phcited by 46

Not on the map. Why?

Abstract (arXiv)

Neural architectures that learn potential energy surfaces from molecular data have undergone fast improvement in recent years. A key driver of this success is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling with system size partly relies upon a spatial distance limit on messages. While this focus on locality is a useful inductive bias, it also impedes the learning of long-range interactions such as electrostatics and van der Waals forces. To address this drawback, we propose Ewald message passing: a nonlocal Fourier space scheme which limits interactions via a cutoff on frequency instead of distance, and is theoretically well-founded in the Ewald summation method. It can serve as an augmentation on top of existing MPNN architectures as it is computationally inexpensive and agnostic to architectural details. We test the approach with four baseline models and two datasets containing diverse periodic (OC20) and aperiodic structures (OE62). We observe robust improvements in energy mean absolute errors across all models and datasets, averaging 10% on OC20 and 16% on OE62. Our analysis shows an outsize impact of these improvements on structures with high long-range contributions to the ground truth energy.

Key facts. Cited by 46 papers in the map (data through 7 Oct 2026). Builds on 26 of its 58 references that are in the map. First posted on arXiv 8 Mar 2023.

Builds on

26 of 58 references.

References · 26

  1. 2017Attention Is All You Needcited by 59,932
  2. 2017Graph Attention Networkscited by 8,854
  3. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  4. 2018Representation Learning on Graphs with Jumping Knowledge Networkscited by 875
  5. 2022Recipe for a General, Powerful, Scalable Graph Transformercited by 702
  6. 2020A Generalization of Transformer Networks to Graphscited by 653
  7. 2020On the Bottleneck of Graph Neural Networks and its Practical Implicationscited by 624
  8. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 569
  9. 2021Rethinking Graph Transformers with Spectral Attentioncited by 468
  10. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 438
  11. 2021GemNet: Universal Directional Graph Neural Networks for Moleculescited by 342
  12. 2021Do Transformers Really Perform Bad for Graph Representation?cited by 318
14 more
  1. 2021Crystal Diffusion Variational Autoencoder for Periodic Material Generationcited by 268
  2. 2019Cormorant: Covariant Molecular Neural Networkscited by 262
  3. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
  4. 2020The Open Catalyst 2020 (OC20) Dataset and Community Challengescited by 246
  5. 2021GraphiT: Encoding Graph Structure in Transformerscited by 146
  6. 2021SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effectscited by 103
  7. 2022The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentialscited by 100
  8. 2020Non-Local Graph Neural Networkscited by 91
  9. 2020On the Universality of Rotation Equivariant Point Cloud Networkscited by 80
  10. 2022GemNet-OC: Developing Graph Neural Networks for Large and Diverse Molecular Simulation Datasetscited by 61
  11. 2020Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structurescited by 44
  12. 2022So3krates: Equivariant attention for interactions on arbitrary length-scales in molecular systemscited by 39
  13. 2021Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functionscited by 36
  14. 2022Capturing long-range interaction with reciprocal space neural networkcited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 14; 2025: 17; 2026: 8.

46 papers.

Sort:

By area:

  1. 2024Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Spacecited by 88
  2. 2023A Hitchhiker's Guide to Geometric GNNs for 3D Atomic Systemscited by 65
  3. 2023Efficient Approximations of Complete Interatomic Potentials for Crystal Property Predictioncited by 37
  4. 2024Latent Ewald summation for machine learning of long-range interactionscited by 26
  5. 2023Navigating protein landscapes with a machine-learned transferable coarse-grained modelcited by 22
  6. 2024Spatio-Spectral Graph Neural Networkscited by 17
  7. 2024Crystalformer: Infinitely Connected Attention for Periodic Structure Encodingcited by 15
  8. 2023Implicit Transfer Operator Learning: Multiple Time-Resolution Surrogates for Molecular Dynamicscited by 14
  9. 2024Euclidean Fast Attention -- Machine Learning Global Atomic Representations at Linear Costcited by 12
  10. 2023Neural Atoms: Propagating Long-range Interaction in Molecular Graphs through Efficient Communication Channelcited by 10
  11. 2024Neural P3M: A Long-Range Interaction Modeling Enhancer for Geometric GNNscited by 9
  12. 2025A universal augmentation framework for long-range electrostatics in machine learning interatomic potentialscited by 9
34 more
  1. 2024Gaussian Plane-Wave Neural Operator for Electron Density Estimationcited by 8
  2. 2024Neural Pfaffians: Solving Many Many-Electron Schr"odinger Equationscited by 8
  3. 2024Learning charges and long-range interactions from energies and forcescited by 7
  4. 2024Learning Equivariant Non-Local Electron Density Functionalscited by 6
  5. 2025chemtrain-deploy: A parallel and scalable framework for machine learning potentials in million-atom MD simulationscited by 6
  6. 2023Learning Interatomic Potentials at Multiple Scalescited by 5
  7. 2025Learning Non-Local Molecular Interactions via Equivariant Local Representations and Charge Equilibrationcited by 5
  8. 2025DualEquiNet: A Dual-Space Hierarchical Equivariant Network for Large Biomoleculescited by 5
  9. 2025HollowFlow: Efficient Sample Likelihood Evaluation using Hollow Message Passingcited by 5
  10. 2025Beyond Structure: Invariant Crystal Property Prediction with Pseudo-Particle Ray Diffractioncited by 4
  11. 2023Equivariant Matrix Function Neural Networkscited by 3
  12. 2024On Representing Electronic Wave Functions with Sign Equivariant Neural Networkscited by 2
  13. 2024SE3Set: Harnessing equivariant hypergraph neural networks for molecular representation learningcited by 2
  14. 2025Machine-Learning Interatomic Potentials for Long-Range Systemscited by 2
  15. 2025Generalization of Long-Range Machine Learning Potentials in Complex Chemical Spacescited by 2
  16. 2026Polarizable atomic multipoles for learning long-range electrostaticscited by 2
  17. 2024Hybrid Quantum Graph Neural Network for Molecular Property Predictioncited by 1
  18. 2025ReciNet: Reciprocal Space-Aware Long-Range Modeling for Crystalline Property Predictioncited by 1
  19. 2025ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamicscited by 1
  20. 2025Hierarchical geometric deep learning enables scalable analysis of molecular dynamicscited by 1
  21. 2026Global Plane Waves From Local Gaussians: Periodic Charge Densities in a Blinkcited by 1
  22. 2026An SO(3)-equivariant reciprocal-space neural potential for long-range interactionscited by 1
  23. 2026Machine Learning Multiscale Interactionscited by 1
  24. 2024Improving Molecular Modeling with Geometric GNNs: an Empirical Studycited by 0
  25. 2025SSLFusion: Scale & Space Aligned Latent Fusion Model for Multimodal 3D Object Detectioncited by 0
  26. 2025Electrostatics from Laplacian Eigenbasis for Neural Network Interatomic Potentialscited by 0
  27. 2025The Generalized Skew Spectrum of Graphscited by 0
  28. 2025MCGM: Multi-stage Clustered Global Modeling for Long-range Interactions in Moleculescited by 0
  29. 2025Reciprocal Space Attention for Learning Long-Range Interactionscited by 0
  30. 2025Long-range electrostatics for machine learning interatomic potentials is easier than we thoughtcited by 0
  31. 2026Physics-Informed Long-Range Coulomb Correction for Machine-learning Hamiltonianscited by 0
  32. 2026Distilling latent electrostatics from foundation machine learning interatomic potentialscited by 0
  33. 2026SupraTITO: Transferable Generative Molecular Dynamics for Supramolecular Systemscited by 0
  34. 2026NEMORA: Neural Equivariant Multipole Operators for Long-Range Atomistic Learningcited by 0