| Hierarchical Graph Pooling with Structure Learning | ✓ Link | 84.91 | | | HGP-SL | 2019-11-14 |
| Randomized Schur Complement Views for Graph Contrastive Learning | ✓ Link | 84.3 | | | rLap (unsupervised) | 2023-06-06 |
| Template based Graph Neural Network with Optimal Transport Distances | ✓ Link | 82.9 | | | TFGW ADJ (L=2) | 2022-05-31 |
| An end-to-end attention-based approach for learning on graphs | ✓ Link | 82.679±0.799 | | | ESA (Edge set attention, no positional encodings) | 2024-02-16 |
| FIT-GNN: Faster Inference Time for GNNs Using Coarsening | ✓ Link | 82.1 | | 0.0016 | FIT-GNN | 2024-10-19 |
| Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning | ✓ Link | 81.70% | | | DUGNN | 2019-09-22 |
| Maximum Entropy Weighted Independent Set Pooling for Graph Neural Networks | ✓ Link | 80.71% | | | MEWISPool | 2021-07-03 |
| CIN++: Enhancing Topological Message Passing | ✓ Link | 80.5 | | | CIN++ | 2023-06-06 |
| Relational Reasoning Over Spatial-Temporal Graphs for Video Summarization | | 80.36% | | | SAEPool | 2022-04-06 |
| Multi-hop Attention-based Graph Pooling: A Personalized PageRank Perspective | ✓ Link | 80.36% | | | MAGPool | 2024-03-04 |
| Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role-based Similarity | ✓ Link | 80.12 ±0.32 | | | UGT | 2023-08-18 |
| Universal Graph Transformer Self-Attention Networks | ✓ Link | 80.01% | | | U2GNN (Unsupervised) | 2019-09-26 |
| Mutual Information Maximization in Graph Neural Networks | ✓ Link | 78.97% | | | sGIN | 2019-05-21 |
| Quantum-based subgraph convolutional neural networks | | 78.80% | | | QS-CNNs (Quantum Walk) | 2019-04-01 |
| Learning metrics for persistence-based summaries and applications for graph classification | ✓ Link | 78.8% | | | WKPI-kmeans | 2019-04-27 |
| Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes | | 78.8% | | | PIN | 2023-08-13 |
| Graph Representation Learning via Hard and Channel-Wise Attention Networks | ✓ Link | 78.65% | | | hGANet | 2019-07-05 |
| Universal Graph Transformer Self-Attention Networks | ✓ Link | 78.53% | | | U2GNN | 2019-09-26 |
| Quantum-based subgraph convolutional neural networks | | 78.35% | | | DS-CNNs (Random Walk) | 2019-04-01 |
| Graph Representation Learning via Hard and Channel-Wise Attention Networks | ✓ Link | 78.23% | | | cGANet | 2019-07-05 |
| Cell Attention Networks | ✓ Link | 78.2% | | | CAN | 2022-09-16 |
| Graph Representation Learning via Hard and Channel-Wise Attention Networks | ✓ Link | 77.92% | | | GANet | 2019-07-05 |
| Graph Star Net for Generalized Multi-Task Learning | ✓ Link | 77.90% | | | GraphStar | 2019-06-21 |
| Relation order histograms as a network embedding tool | ✓ Link | 77.89% | | | NERO | 2021-06-09 |
| Fea2Fea: Exploring Structural Feature Correlations via Graph Neural Networks | ✓ Link | 77.8% | | | Fea2Fea-s2 | 2021-06-24 |
| Discriminative Graph Autoencoder | | 77.71% | | | DGA | 2018-11-17 |
| Graph U-Nets | ✓ Link | 77.68% | | | Graph U-Nets | 2019-05-11 |
| How Attentive are Graph Attention Networks? | ✓ Link | 77.679±2.187 | | | GATv2 | 2021-05-30 |
| Principal Neighbourhood Aggregation for Graph Nets | ✓ Link | 77.679±3.281 | | | PNA | 2020-04-12 |
| Gaussian-Induced Convolution for Graphs | | 77.65% | | | GIC | 2018-11-11 |
| Graph isomorphism UNet | ✓ Link | 77.6% | | | GIUNet | 2023-08-23 |
| Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification | ✓ Link | 77.44% | | | GFN-light | 2019-05-11 |
| Semi-Supervised Graph Classification: A Hierarchical Graph Perspective | ✓ Link | 77.26% | | | SEAL-SAGE | 2019-04-10 |
| Provably Powerful Graph Networks | ✓ Link | 77.20% | | | PPGN | 2019-05-27 |
| Recipe for a General, Powerful, Scalable Graph Transformer | ✓ Link | 77.143±1.494 | | | GraphGPS | 2022-05-25 |
| Understanding Attention and Generalization in Graph Neural Networks | ✓ Link | 77.09% | | | Weak-supervised ChebyNet | 2019-05-08 |
| Segmented Graph-Bert for Graph Instance Modeling | ✓ Link | 77.09% | | | SEG-BERT | 2020-02-09 |
| Improving Attention Mechanism in Graph Neural Networks via Cardinality Preservation | ✓ Link | 76.81% | | | GAT-GC (f-Scaled) | 2019-07-04 |
| Graph Attention Networks | ✓ Link | 76.786±1.670 | | | GAT | 2017-10-30 |
| Subgraph Networks with Application to Structural Feature Space Expansion | | 76.78% | | | Deep WL SGN(0,1,2) | 2019-03-21 |
| Graph Convolutional Networks with EigenPooling | ✓ Link | 76.60% | | | EigenGCN-3 | 2019-04-30 |
| Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules | ✓ Link | 76.5% | | | Multigraph ChebNet | 2018-11-23 |
| Wasserstein Embedding for Graph Learning | ✓ Link | 76.5% | | | WEGL | 2020-06-16 |
| A Novel Higher-order Weisfeiler-Lehman Graph Convolution | ✓ Link | 76.5 | | | 2-WL-GNN | 2020-07-01 |
| Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification | ✓ Link | 76.46% | | | GFN | 2019-05-11 |
| Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks | ✓ Link | 76.4% | | | Shortest-Path Kernel | 2018-10-04 |
| Graph Capsule Convolutional Neural Networks | ✓ Link | 76.40% | | | GCAPS-CNN | 2018-05-21 |
| On Valid Optimal Assignment Kernels and Applications to Graph Classification | | 76.4% | | | WL-OA | 2016-06-03 |
| The Multiscale Laplacian Graph Kernel | | 76.34% | | | MLG | 2016-03-20 |
| DAGCN: Dual Attention Graph Convolutional Networks | ✓ Link | 76.33% | | | DAGCN | 2019-04-04 |
| Graph Kernels: A Survey | | 76.31% | | | CORE-SP | 2019-04-27 |
| DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks | ✓ Link | 76.3% | | | DropGIN | 2021-11-11 |
| Capsule Graph Neural Network | ✓ Link | 76.28% | | | CapsGNN | 2019-05-01 |
| An End-to-End Deep Learning Architecture for Graph Classification | ✓ Link | 76.26% | | | DGCNN | 2018-04-29 |
| Hierarchical Graph Representation Learning with Differentiable Pooling | ✓ Link | 76.25% | | | GNN (DiffPool) | 2018-06-22 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 76.17 | | | R-GIN + PANDA | 2024-06-06 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 76 | | | GCN + PANDA | 2024-06-06 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 76 | | | R-GCN + PANDA | 2024-06-06 |
| Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks | ✓ Link | 75.9% | | | k-GNN | 2018-10-04 |
| PANDA: Expanded Width-Aware Message Passing Beyond Rewiring | ✓ Link | 75.759 | | | GIN + PANDA | 2024-06-06 |
| Deep Graph Kernels | | 75.68% | | | DGK | 2015-08-10 |
| Graph-level Representation Learning with Joint-Embedding Predictive Architectures | ✓ Link | 75.67% | | | Graph-JEPA | 2023-09-27 |
| TREE-G: Decision Trees Contesting Graph Neural Networks | ✓ Link | 75.6 | | | TREE-G | 2022-07-06 |
| Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 75.536±1.622 | | | GCN | 2016-09-09 |
| How Powerful are Graph Neural Networks? | ✓ Link | 75.536±1.851 | | | GIN | 2018-10-01 |
| DiffWire: Inductive Graph Rewiring via the Lovász Bound | ✓ Link | 75.38% | | | CT-Layer | 2022-06-15 |
| A Persistent Weisfeiler–Lehman Procedure for Graph Classification | ✓ Link | 75.36% | | | P-WL-UC | 2019-06-09 |
| DiffWire: Inductive Graph Rewiring via the Lovász Bound | ✓ Link | 75.34% | | | GAP-Layer (Ncut) | 2022-06-15 |
| Fast Graph Representation Learning with PyTorch Geometric | ✓ Link | 75.1% | | | DiffPool | 2019-03-06 |
| DGCNN: Disordered Graph Convolutional Neural Network Based on the Gaussian Mixture Model | | 75.1% | | | DGCNN | 2017-12-10 |
| Accurate Learning of Graph Representations with Graph Multiset Pooling | ✓ Link | 75.09% | | | GMT | 2021-02-23 |
| DiffWire: Inductive Graph Rewiring via the Lovász Bound | ✓ Link | 75.03% | | | GAP-Layer (Rcut) | 2022-06-15 |
| PiNet: A Permutation Invariant Graph Neural Network for Graph Classification | ✓ Link | 75% | | | PiNet (Learned p and q) | 2019-05-08 |
| DiffWire: Inductive Graph Rewiring via the Lovász Bound | ✓ Link | 74.91% | | | DiffWire | 2022-06-15 |
| Online Graph Dictionary Learning | ✓ Link | 74.86 | | | GDL-g (SP) | 2021-02-12 |
| Variational Recurrent Neural Networks for Graph Classification | ✓ Link | 74.8% | | | VRGC | 2019-02-07 |
| A simple yet effective baseline for non-attributed graph classification | ✓ Link | 74.7% | | | LDP + distance | 2018-11-08 |
| Weisfeiler and Leman go sparse: Towards scalable higher-order graph embeddings | ✓ Link | 74.60% | | | δ-2-LWL | 2019-04-02 |
| Optimal Transport for structured data with application on graphs | ✓ Link | 74.55% | | | FGW sp | 2018-05-23 |
| Wasserstein Weisfeiler-Lehman Graph Kernels | ✓ Link | 74.28% | | | WWL | 2019-06-04 |
| Distinguishing Enzyme Structures from Non-enzymes Without Alignments00628-4) | | 74.22% | | | RW | 2003-07-18 |
| ASAP: Adaptive Structure Aware Pooling for Learning Hierarchical Graph Representations | ✓ Link | 74.19% | | | ASAP | 2019-11-18 |
| Capsule Neural Networks for Graph Classification using Explicit Tensorial Graph Representations | | 74.1% | | | BC + Capsules | 2019-02-22 |
| A simple yet effective baseline for non-attributed graph classification | ✓ Link | 73.7% | | | LDP + Labels | 2018-11-08 |
| A Fair Comparison of Graph Neural Networks for Graph Classification | ✓ Link | 73.7% | | | DiffPool | 2019-12-20 |
| A Simple Baseline Algorithm for Graph Classification | ✓ Link | 73.6% | | | SF + RFC | 2018-10-22 |
| Edge Contraction Pooling for Graph Neural Networks | | 73.5% | | | EdgePool w GraphSAGE | 2019-05-27 |
| Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling | ✓ Link | 73.3% | | | Graph2Vec | 2019-10-24 |
| graph2vec: Learning Distributed Representations of Graphs | ✓ Link | 73.3% ± 2.05% | | | graph2vec | 2017-07-17 |
| A Fair Comparison of Graph Neural Networks for Graph Classification | ✓ Link | 73% | | | GraphSAGE | 2019-12-20 |
| Function Space Pooling For Graph Convolutional Networks | | 72.8% | | | Function Space Pooling | 2019-05-15 |
| Strengthening structural baselines for graph classification using Local Topological Profile | ✓ Link | 72.7 ± 4.2 | 72.7 ± 4.2 | | Local Topological Profile (LTP) | 2023-05-01 |
| A simple yet effective baseline for non-attributed graph classification | ✓ Link | 72.7% | | | LDP | 2018-11-08 |
| Edge Contraction Pooling for Graph Neural Networks | | 72.5% | | | EdgePool | 2019-05-27 |
| A Non-Negative Factorization approach to node pooling in Graph Convolutional Neural Networks | | 72.1% | | | 1-NMFPool | 2019-09-07 |
| SPI-GCN: A Simple Permutation-Invariant Graph Convolutional Network | | 72.06% | | | SPI-GCN | 2019-04-08 |
| Self-Attention Graph Pooling | ✓ Link | 71.86% | | | SAGPool_h | 2019-04-17 |
| Efficient graphlet kernels for large graph comparison | | 71.67% | | | GK | 2009-04-16 |
| Rep the Set: Neural Networks for Learning Set Representations | ✓ Link | 70.74% | | | ApproxRepSet | 2019-04-03 |
| Self-Attention Graph Pooling | ✓ Link | 70.04% | | | SAGPool_g | 2019-04-17 |
| 0/1 Deep Neural Networks via Block Coordinate Descent | | 65.7% ± 4.2% | | | Eff.resistance graph kernel | 2022-06-19 |
| Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns | ✓ Link | | 72.05 | | G-Tuning | 2023-12-21 |
| How Powerful are Graph Neural Networks? | ✓ Link | 76,2% | | | GIN-0 | 2018-10-01 |