Research map / Graph learning / GNN architectures
Equivariant networks for molecules
Predict molecular and crystal properties with equivariant message passing.
43 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in GNN architectures, keyword cluster “message passing · equivariant · mpnns”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
12 papers in 2023–24 and 18 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 2.5% to 3.5%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2017 · Neural Message Passing for Quantum Chemistry (ICML 2017 · 3,844 citations)
- 2020 · Learning to Simulate Complex Physics with Graph Networks (ICML 2020 · 434 citations)
- 2021 · Equivariant message passing for the prediction of tensorial properties and molecular spectra (ICML 2021 · 253 citations)
- 2022 · ComENet: Towards Complete and Efficient Message Passing for 3D Molecular Graphs (NeurIPS 2022 · 27 citations)
- 2023 · Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs (ICLR 2023 · 67 citations)
- 2024 · HAGO-Net: Hierarchical Geometric Message Passing for Molecular Representation Learning (AAAI 2024 · 8 citations)
- 2025 · AeroGTO: An Efficient Graph-Transformer Operator for Learning Large-Scale Aerodynamics of 3D Vehicle Geometries (AAAI 2025 · 6 citations)
- 2026 · Equivariant Spherical Transformer for Efficient Molecular Modeling (NeurIPS 2026 · 1 citation)
Earliest papers here
- Neural Message Passing for Quantum Chemistry (ICML 2017 · 3,844 citations)
- Learning to Simulate Complex Physics with Graph Networks (ICML 2020 · 434 citations)
- Directional Message Passing for Molecular Graphs (ICLR 2020 · 349 citations)
What has been tried
25 papers here report 299 measured changes (a component added, removed or swapped, compared with a baseline): helped 70%, no clear effect 7%, hurt 20%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add sinusoidal positional encoding (SPE) to GNN/decoder | Ahmed-Body · L2 relative error | 0.0860 -> 0.0757 | Helped | AeroGTO: An Efficient Graph-Transformer Operator for Learning Large-Scale Aerodynamics of 3D Vehicle Geometries (AAAI 2025 · 6 citations) |
| add the long-range component | AT-AT-CG-CG · energy MAE (kcal/mol) | 0.1563 -> 0.1064 | Helped | Long-Short-Range Message-Passing: A Physics-Informed Framework to Capture Non-Local Interaction for Scalable Molecular Dynamics Simulation (ICLR 2024 · 6 citations) |
| Use j-side rotations only instead of rotations from both sides | MD17 Aspirin and Benzene · force MAE | 0.316 -> 0.323; 0.151 -> 0.152 | Mixed | A Plug-and-Play Quaternion Message-Passing Module for Molecular Conformation Representation (AAAI 2024 · 3 citations) |
| add anisotropic intra-level propagation | CylinderFlow, Flag · RMSE-All (×10⁻²) | 16.98 -> 9.621; 168.1 -> 121.1 | Helped | EvoMesh: Adaptive Physical Simulation with Hierarchical Graph Evolutions (ICML 2025 · 0 citations) |
| add CSIP instead of standard mean pooling | CovidVaccine · RMSE | 0.356 ±0.002 -> 0.353±0.006 | Helped | DualEqui: A Dual-Space Hierarchical Equivariant Network for Large Biomolecules (NeurIPS 2025 · 0 citations) |
| embed proportions in atomic features rather than apply scalar weighting to molecule embeddings | CALiSol · MSE | 3.657 -> 0.552 | Helped | Geometric Mixture Models for Electrolyte Conductivity Prediction (NeurIPS 2025 · 0 citations) |
| Replace SymPE with standard-normal noise | QM9 explicit hydrogens · NLL | 30.3 -> 126.5 | Hurt | Improving Equivariant Networks with Probabilistic Symmetry Breaking (ICLR 2025 · 0 citations) |
| Share simplicial message-passing networks instead of using separate networks | MD17 Aspirin · Seconds/Step | 0.37 -> 0.22 | Helped | Clifford Group Equivariant Simplicial Message Passing Networks (ICLR 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 10 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- QM9 · 2 of 10 papers
Baselines
- SchNet · 2 of 10 papers
- TFN · 2 of 10 papers
- MLP · 2 of 10 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 30 papers; from OpenAlex and the papers' affiliations).
- Fudan University · 3 papers
- Beijing Institute of Technology · 2 papers
- Renmin University of China · 2 papers
- Xi’an Jiaotong University · 2 papers
- Shanghai Jiao Tong University · 2 papers
- Tsinghua University · 2 papers
- Nanjing University · 2 papers
- Microsoft · 2 papers
- Texas A&M · 2 papers
- Texas A&M University · 2 papers
Use it
Download BibTeX of 35 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Expressive equivariant message passing (109 papers)
- Graph rewiring and scalable message passing (72 papers)
- Structural GNN variants (65 papers)
Elsewhere in the topic
- Graph structure learning (201 papers)
- GNN generalization theory (187 papers)
- Simple and scalable GNNs (131 papers)
- Robustness, privacy, and fairness of GNNs (94 papers)
- Theory of graph convolutional networks (83 papers)
- Deep graph convolution and simplification (82 papers)
- Physics-inspired GNNs (81 papers)
- Weisfeiler-Leman expressivity hierarchies (66 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).