mlsanityData to 7 Oct 2026

Equivariant Graph Attention Networks for Molecular Property Prediction

arXiv 2202.09891 ↗First posted 20 Feb 2022cs.LG q-bio.BMcited by 19

Not on the map. Why?

Abstract (arXiv)

Learning and reasoning about 3D molecular structures with varying size is an emerging and important challenge in machine learning and especially in drug discovery. Equivariant Graph Neural Networks (GNNs) can simultaneously leverage the geometric and relational detail of the problem domain and are known to learn expressive representations through the propagation of information between nodes leveraging higher-order representations to faithfully express the geometry of the data, such as directionality in their intermediate layers. In this work, we propose an equivariant GNN that operates with Cartesian coordinates to incorporate directionality and we implement a novel attention mechanism, acting as a content and spatial dependent filter when propagating information between nodes. We demonstrate the efficacy of our architecture on predicting quantum mechanical properties of small molecules and its benefit on problems that concern macromolecular structures such as protein complexes.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Builds on 25 of its 36 references that are in the map. First posted on arXiv 20 Feb 2022.

Builds on

25 of 36 references.

References · 25

  1. 2017Attention Is All You Needcited by 59,932
  2. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  3. 2019Transformer-XL: Attentive Language Models Beyond a Fixed-Length Contextcited by 1,830
  4. 2018Relational inductive biases, deep learning, and graph networkscited by 1,341
  5. 2021Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gaugescited by 1,131
  6. 2021E(n) Equivariant Graph Neural Networkscited by 951
  7. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
  8. 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 701
  9. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 602
  10. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 569
  11. 2020Directional Message Passing for Molecular Graphscited by 556
  12. 2019Stand-Alone Self-Attention in Vision Modelscited by 542
13 more
  1. 2021Rethinking Graph Transformers with Spectral Attentioncited by 468
  2. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 438
  3. 2021Do Transformers Really Perform Bad for Graph Representation?cited by 318
  4. 2020Exploring Self-attention for Image Recognitioncited by 313
  5. 2020Learning from Protein Structure with Geometric Vector Perceptronscited by 308
  6. 2019Cormorant: Covariant Molecular Neural Networkscited by 262
  7. 2020Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Datacited by 248
  8. 2019Graph Matching Networks for Learning the Similarity of Graph Structured Objectscited by 202
  9. 2017Learning Steerable Filters for Rotation Equivariant CNNscited by 193
  10. 2022TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentialscited by 179
  11. 2021SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effectscited by 103
  12. 2021Equivariant Graph Neural Networks for 3D Macromolecular Structurecited by 53
  13. 2020Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Propertiescited by 51

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 5; 2024: 3; 2025: 6; 2026: 3.

19 papers.

Sort:

By area:

  1. 2022Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphscited by 258
  2. 2023EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representationscited by 201
  3. 2023A new perspective on building efficient and expressive 3D equivariant graph neural networkscited by 46
  4. 2023Evaluation of the MACE Force Field Architecture: from Medicinal Chemistry to Materials Sciencecited by 25
  5. 2024Generalizing Denoising to Non-Equilibrium Structures Improves Equivariant Force Fieldscited by 18
  6. 2022DPA-1: Pretraining of Attention-based Deep Potential Model for Molecular Simulationcited by 15
  7. 2024Triplet Interaction Improves Graph Transformers: Accurate Molecular Graph Learning with Triplet Graph Transformerscited by 12
  8. 2025Platonic Transformers: A Solid Choice For Equivariancecited by 9
  9. 2026EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformerscited by 7
  10. 2025Diffusion Generative Modeling on Lie Group Representationscited by 6
  11. 2023GPS++: Reviving the Art of Message Passing for Molecular Property Predictioncited by 5
  12. 2023Molecular Geometry-aware Transformer for accurate 3D Atomic System modelingcited by 4
7 more
  1. 2025Equivariant Spherical Transformer for Efficient Molecular Modelingcited by 3
  2. 2024Analysis of Atom-level pretraining with Quantum Mechanics (QM) data for Graph Neural Networks Molecular property modelscited by 1
  3. 2025Potential Score Matching: Debiasing Molecular Structure Sampling with Potential Energy Guidancecited by 0
  4. 2025Constructing 3D Rotational Invariance and Equivariance with Symmetric Tensor Networkscited by 0
  5. 2025Spectral Analysis of Molecular Features: When Richer Features Do Not Guarantee Better Generalizationcited by 0
  6. 2026The Convergence Frontier: Integrating Machine Learning and High Performance Quantum Computing for Next-Generation Drug Discoverycited by 0
  7. 2026Equivariant Graph Neural Networks Improve Optical Spectra Prediction for Materials Screeningcited by 0