| Universal Graph Transformer Self-Attention Networks | ✓ Link | 96.41% | | | U2GNN (Unsupervised) | 2019-09-26 |
| An end-to-end attention-based approach for learning on graphs | ✓ Link | 86.250±0.957 | | | ESA (Edge set attention, no positional encodings) | 2024-02-16 |
| Graph Attention Networks | ✓ Link | 84.250±2.062 | | | GAT | 2017-10-30 |
| Maximum Entropy Weighted Independent Set Pooling for Graph Neural Networks | ✓ Link | 82.13% | | | MEWISPool | 2021-07-03 |
| How Powerful are Graph Neural Networks? | ✓ Link | 81.250±3.775 | | | GIN | 2018-10-01 |
| Pure Transformers are Powerful Graph Learners | ✓ Link | 80.250±3.304 | | | TokenGT | 2022-07-06 |
| How Attentive are Graph Attention Networks? | ✓ Link | 80.000±2.739 | | | GATv2 | 2021-05-30 |
| When Work Matters: Transforming Classical Network Structures to Graph CNN | | 79.90% | | | G_ResNet | 2018-07-07 |
| Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 79.500±3.109 | | | GCN | 2016-09-09 |
| Recipe for a General, Powerful, Scalable Graph Transformer | ✓ Link | 79.250±3.096 | | | GraphGPS | 2022-05-25 |
| Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning | ✓ Link | 78.70% | | | DUGNN | 2019-09-22 |
| Template based Graph Neural Network with Optimal Transport Distances | ✓ Link | 78.3% | | | TFGW ADJ (L=2) | 2022-05-31 |
| Principal Neighbourhood Aggregation for Graph Nets | ✓ Link | 78.000±3.808 | | | PNA | 2020-04-12 |
| Mutual Information Maximization in Graph Neural Networks | ✓ Link | 77.94% | | | sGIN | 2019-05-21 |
| Do Transformers Really Perform Bad for Graph Representation? | ✓ Link | 77.500±2.646 | | | Graphormer | 2021-06-09 |
| Segmented Graph-Bert for Graph Instance Modeling | ✓ Link | 77.2% | | | SEG-BERT | 2020-02-09 |
| Universal Graph Transformer Self-Attention Networks | ✓ Link | 77.04% | | | U2GNN | 2019-09-26 |
| Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes | | 76.6% | | | PIN | 2023-08-13 |
| Graph isomorphism UNet | ✓ Link | 76% | | | GIUNet | 2023-08-23 |
| Subgraph Networks with Application to Structural Feature Space Expansion | | 75.70% | | | Deep WL SGN(0,1,2) | 2019-03-21 |
| DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks | ✓ Link | 75.7% | | | DropGIN | 2021-11-11 |
| Learning metrics for persistence-based summaries and applications for graph classification | ✓ Link | 75.4% | | | WKPI-kcenters | 2019-04-27 |
| Wasserstein Embedding for Graph Learning | ✓ Link | 75.4% | | | WEGL | 2020-06-16 |
| Factorizable Graph Convolutional Networks | ✓ Link | 75.3% | 75.3% | | FactorGCN | 2020-10-12 |
| How Powerful are Graph Neural Networks? | ✓ Link | 75.1% | | | GIN-0 | 2018-10-01 |
| Strengthening structural baselines for graph classification using Local Topological Profile | ✓ Link | 74.5 ± 4.3 | 74.5 ± 4.3 | | Local Topological Profile (LTP) | 2023-05-01 |
| Anonymous Walk Embeddings | ✓ Link | 74.45% | | | AWE | 2018-05-30 |
| Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks | ✓ Link | 74.2% | | | k-GNN | 2018-10-04 |
| Graph-level Representation Learning with Joint-Embedding Predictive Architectures | ✓ Link | 73.68% | | | Graph-JEPA | 2023-09-27 |
| Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks | ✓ Link | 73.5% | | | 3-WL Kernel | 2018-10-04 |
| Accurate Learning of Graph Representations with Graph Multiset Pooling | ✓ Link | 73.48% | | | GMT | 2021-02-23 |
| Weisfeiler and Leman go sparse: Towards scalable higher-order graph embeddings | ✓ Link | 73.4% | | | δ-2-LWL | 2019-04-02 |
| Capsule Graph Neural Network | ✓ Link | 73.10% | | | CapsGNN | 2019-05-01 |
| InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization | ✓ Link | 73.03% | | | InfoGraph | 2019-07-31 |
| Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification | ✓ Link | 73.00% | | | GFN | 2019-05-11 |
| Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification | ✓ Link | 73.00% | | | GFN-light | 2019-05-11 |
| TREE-G: Decision Trees Contesting Graph Neural Networks | ✓ Link | 73% | | | TREE-G | 2022-07-06 |
| Fast Graph Representation Learning with PyTorch Geometric | ✓ Link | 72.8% | | | GIN-0 | 2019-03-06 |
| Provably Powerful Graph Networks | ✓ Link | 72.6% | | | PPGN | 2019-05-27 |
| A Novel Higher-order Weisfeiler-Lehman Graph Convolution | ✓ Link | 72.2 | | | 2-WL-GNN | 2020-07-01 |
| Online Graph Dictionary Learning | ✓ Link | 72.06% | | 51.64 | GDL | 2021-02-12 |
| Graph Capsule Convolutional Neural Networks | ✓ Link | 71.69% | | | GCAPS-CNN | 2018-05-21 |
| Rep the Set: Neural Networks for Learning Set Representations | ✓ Link | 71.46% | | | ApproxRepSet | 2019-04-03 |
| Learning Convolutional Neural Networks for Graphs | ✓ Link | 71.00% | | | PSCN | 2016-05-17 |
| Graph Classification with 2D Convolutional Neural Networks | | 70.40% | | | 2D CNN | 2017-07-29 |
| An End-to-End Deep Learning Architecture for Graph Classification | ✓ Link | 70.03% | | | DGCNN | 2018-04-29 |
| A Fair Comparison of Graph Neural Networks for Graph Classification | ✓ Link | 68.8% | | | GraphSAGE | 2019-12-20 |
| Deep Graph Kernels | | 66.96% | | | DGK | 2015-08-10 |
| SPI-GCN: A Simple Permutation-Invariant Graph Convolutional Network | | 60.40% | | | SPI-GCN | 2019-04-08 |
| An End-to-End Deep Learning Architecture for Graph Classification | ✓ Link | 51.69% | | | DGCNN (sum) | 2018-04-29 |
| Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns | ✓ Link | | 74.30 | | G-Tuning | 2023-12-21 |