| An end-to-end attention-based approach for learning on graphs | ✓ Link | 79.423±1.658 | | ESA (Edge set attention, no positional encodings) | 2024-02-16 |
| Recipe for a General, Powerful, Scalable Graph Transformer | ✓ Link | 78.667±4.625 | | GraphGPS | 2022-05-25 |
| Graph Attention Networks | ✓ Link | 78.611±1.556 | | GAT | 2017-10-30 |
| Bridging the Gap Between Spectral and Spatial Domains in Graph Neural Networks | ✓ Link | 78.39 | | DSGCN-allfeat | 2020-03-26 |
| How Attentive are Graph Attention Networks? | ✓ Link | 77.987±2.112 | | GATv2 | 2021-05-30 |
| Template based Graph Neural Network with Optimal Transport Distances | ✓ Link | 75.1 | | TFGW SP (L=2) | 2022-05-31 |
| Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 73.466±4.372 | | GCN | 2016-09-09 |
| A New Perspective on the Effects of Spectrum in Graph Neural Networks | ✓ Link | 73.33 | | Norm-GN | 2021-12-14 |
| Principal Neighbourhood Aggregation for Graph Nets | ✓ Link | 73.021±2.512 | | PNA | 2020-04-12 |
| Online Graph Dictionary Learning | ✓ Link | 71.47 | | GDL-g (SP) | 2021-02-12 |
| Optimal Transport for structured data with application on graphs | ✓ Link | 71.00% | | FGW sp | 2018-05-23 |
| Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification | ✓ Link | 70.17% | | GFN | 2019-05-11 |
| Graph isomorphism UNet | ✓ Link | 70% | | GIUNet | 2023-08-23 |
| Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification | ✓ Link | 69.50% | | GFN-light | 2019-05-11 |
| Hierarchical Graph Pooling with Structure Learning | ✓ Link | 68.79 | | HGP-SL | 2019-11-14 |
| How Powerful are Graph Neural Networks? | ✓ Link | 68.303±4.170 | | GIN | 2018-10-01 |
| When Work Matters: Transforming Classical Network Structures to Graph CNN | | 67.50% | | G_Inception | 2018-07-07 |
| Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning | ✓ Link | 67.30% | | DUGNN | 2019-09-22 |
| Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role-based Similarity | ✓ Link | 67.22±3.92 | | UGT | 2023-08-18 |
| Graph Star Net for Generalized Multi-Task Learning | ✓ Link | 67.1% | | GraphStar | 2019-06-21 |
| Bridging the Gap Between Spectral and Spatial Domains in Graph Neural Networks | ✓ Link | 65.13 | | DSGCN-nodelabel | 2020-03-26 |
| DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks | ✓ Link | 65.128±4.117 | | DropGIN | 2021-11-11 |
| Graph Convolutional Networks with EigenPooling | ✓ Link | 65.0% | | EigenGCN-3 | 2019-04-30 |
| Hierarchical Graph Representation Learning with Differentiable Pooling | ✓ Link | 63.33% | | S2V (with 2 DiffPool) | 2018-06-22 |
| Hierarchical Graph Representation Learning with Differentiable Pooling | ✓ Link | 62.53% | | GNN (DiffPool) | 2018-06-22 |
| Gaussian-Induced Convolution for Graphs | | 62.50% | | GIC | 2018-11-11 |
| Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules | ✓ Link | 61.7% | | Multigraph ChebNet | 2018-11-23 |
| Wasserstein Embedding for Graph Learning | ✓ Link | 60.5 | | WEGL | 2020-06-16 |
| A Fair Comparison of Graph Neural Networks for Graph Classification | ✓ Link | 59.6% | | GIN | 2019-12-20 |
| TREE-G: Decision Trees Contesting Graph Neural Networks | ✓ Link | 59.6 | | TREE-G | 2022-07-06 |
| Wasserstein Weisfeiler-Lehman Graph Kernels | ✓ Link | 59.13% | | WWL | 2019-06-04 |
| Improving Attention Mechanism in Graph Neural Networks via Cardinality Preservation | ✓ Link | 58.45 | | GAT-GC (f-Scaled) | 2019-07-04 |
| Weisfeiler and Leman go sparse: Towards scalable higher-order graph embeddings | ✓ Link | 58.2% | | δ-2-LWL | 2019-04-02 |
| A Fair Comparison of Graph Neural Networks for Graph Classification | ✓ Link | 58.2% | | GraphSAGE | 2019-12-20 |
| DAGCN: Dual Attention Graph Convolutional Networks | ✓ Link | 58.17% | | DAGCN | 2019-04-04 |
| Evolution of Graph Classifiers | ✓ Link | 55.67 | | Evolution of Graph Classifiers | 2019-10-04 |
| Capsule Graph Neural Network | ✓ Link | 54.67% | | CapsGNN | 2019-05-01 |
| Deep Graph Kernels | | 53.43% | | DGK | 2015-08-10 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 53.1 | | R-GIN + PANDA | 2024-06-06 |
| Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs | ✓ Link | 52.67% | | ECC (5 scores) | 2017-04-10 |
| SPI-GCN: A Simple Permutation-Invariant Graph Convolutional Network | | 50.17% | | SPI-GCN | 2019-04-08 |
| Fea2Fea: Exploring Structural Feature Correlations via Graph Neural Networks | ✓ Link | 48.5 | | Fea2Fea-s2 | 2021-06-24 |
| Variational Recurrent Neural Networks for Graph Classification | ✓ Link | 48.4% | | VRGC | 2019-02-07 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 46.2 | | GIN + PANDA | 2024-06-06 |
| Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling | ✓ Link | 43.9% | | NDP | 2019-10-24 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 43.9 | | R-GCN + PANDA | 2024-06-06 |
| A Simple Baseline Algorithm for Graph Classification | ✓ Link | 43.7% | | SF + RFC | 2018-10-22 |
| Strengthening structural baselines for graph classification using Local Topological Profile | ✓ Link | 42.5 ± 4.1 | 42.5 ± 4.1 | Local Topological Profile (LTP) | 2023-05-01 |
| A simple yet effective baseline for non-attributed graph classification | ✓ Link | 35.3% | | LDP | 2018-11-08 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 31.55 | | GCN + PANDA | 2024-06-06 |
| DEMO-Net: Degree-specific Graph Neural Networks for Node and Graph Classification | ✓ Link | 27.2 | | DEMO-Net(weight) | 2019-06-05 |
| Capsule Neural Networks for Graph Classification using Explicit Tensorial Graph Representations | | 27% | | BC + Capsules | 2019-02-22 |
| A Non-Negative Factorization approach to node pooling in Graph Convolutional Neural Networks | | 24.1% | | 1-NMFPool | 2019-09-07 |
| Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns | ✓ Link | | 26.70 | G-Tuning | 2023-12-21 |