Graph Classification on IMDB-B (Accuracy and Std Dev)
80.5Mean AccuracySNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SNNValidation=Tenfold cross-validation, Variant=SNN(α, (1, 1)), γ=0.52024.12 | 80.5 | — | |
| TL-GNNValidation=Tenfold cross-validation2024.12 | 79.7 | — | |
| CCIN2026.05 | 78.3 | — | |
| GSNValidation=Tenfold cross-validation2024.12 | 77.8 | — | |
| GSNEvaluation Protocol=Standard eval.2026.05 | 77.8 | — | |
| ShareGNNEvaluation Protocol=Standard eval.2026.05 | 77.7 | — | |
| DiGGRLearning Paradigm=Unsupervised2024.08 | 77.68 | — | |
| ALL-INTransfer Learning=Frozen encoder with lightweight classifier2026.05 | 77.2 | — | |
| GNTKCategory=Graph Kernel Methods, Validation=Tenfold cross-validation2024.12 | 76.9 | — | |
| G2N2Peak GPU Memory (MB)=8462026.05 | 76.8 | — | |
| VEPMLearning Paradigm=Supervised/Semi-supervised2024.08 | 76.7 | — | |
| DisenSemiLearning Paradigm=Supervised/Semi-supervised2024.08 | 76.7 | — | |
| PINEvaluation Protocol=Standard eval.2026.05 | 76.6 | — | |
| CTQWformer2026.05 | 76.4 | — | |
| TopoTune2026.05 | 76.3 | — | |
| GraphGPS2026.05 | 76.3 | — | |
| DGCLLearning Paradigm=Unsupervised2024.08 | 75.9 | — | |
| MGNNIValidation=10-fold cross-validation, Setting=Inductive2022.10 | 75.8 | 3.4 | |
| CH-CL2026.04 | 75.62 | — | |
| SINValidation=Tenfold cross-validation2024.12 | 75.6 | — | |
| CINValidation=Tenfold cross-validation2024.12 | 75.6 | — | |
| GraphGPS2026.05 | 75.6 | — | |
| CINEvaluation Protocol=Standard eval.2026.05 | 75.6 | — | |
| SINEvaluation Protocol=Standard eval.2026.05 | 75.6 | — | |
| AUG-MAEApproach Category=Feature-based2024.10 | 75.6 | — | |
| CINPeak GPU Memory (MB)=4542026.05 | 75.6 | — | |
| GraphMAELearning Paradigm=Unsupervised2024.08 | 75.52 | — | |
| GraphMAE2Approach Category=Feature-based2024.10 | 75.5 | — | |
| HGNNPeak GPU Memory (MB)=192026.05 | 75.5 | — | |
| GMCL2026.04 | 75.48 | — | |
| GCBM-E2026.03 | 75.4 | — | |
| GRIT2026.05 | 75.4 | — | |
| GINValidation=10-fold cross-validation, Setting=Inductive2022.10 | 75.1 | 5.1 | |
| GINInput node features=node degrees (one-hot)2019.05 | 75.1 | 5.1 | |
| GINValidation=Tenfold cross-validation2024.12 | 75.1 | — | |
| GIN2026.05 | 75.1 | — | |
| WKPIConfig Category=Top NT, Landmarks (K)=200, Evaluation Protocol=5 seeds x 10 folds2026.05 | 75.1 | — | |
| GIN2026.05 | 75.1 | — | |
| GINLearning Paradigm=Supervised2024.08 | 75.1 | — | |
| InfoGCLLearning Paradigm=Unsupervised2024.08 | 75.1 | — | |
| GIN-02026.05 | 75.1 | — | |
| GAEApproach Category=Structure-based2024.10 | 75.1 | — | |
| GraphMAEApproach Category=Feature-based2024.10 | 75 | — | |
| sCWNcPeak GPU Memory (MB)=192026.05 | 75 | — | |
| Graphormer2026.05 | 74.9 | — | |
| GRDL2026.03 | 74.8 | — | |
| Natural GN2026.05 | 74.8 | — | |
| GraphSNN2026.05 | 74.8 | — | |
| GATPeak GPU Memory (MB)=262026.05 | 74.8 | — | |
| TopoGCL2026.04 | 74.67 | — | |
| GAWL2026.03 | 74.6 | — | |
| sCWNrPeak GPU Memory (MB)=1392026.05 | 74.6 | — | |
| KGWL2026.05 | 74.4 | — | |
| GAEfApproach Category=Feature-based2024.10 | 74.4 | — | |
| MaskGAEApproach Category=Structure-based2024.10 | 74.4 | — | |
| WLEvaluation Protocol=Standard eval.2026.05 | 74.3 | — | |
| GCNPeak GPU Memory (MB)=192026.05 | 74.3 | — | |
| SAGEConvPeak GPU Memory (MB)=192026.05 | 74.3 | — | |
| GCBM2026.03 | 74.2 | — | |
| MVGRLLearning Paradigm=Unsupervised2024.08 | 74.2 | — | |
| LGNNInput node features=identical features2019.05 | 74.1 | 4.6 | |
| SEP2026.03 | 74.1 | — | |
| lrGAE 6Approach Category=Structure-based2024.10 | 74 | — | |
| GraphMAE2Learning Paradigm=Unsupervised2024.08 | 73.88 | — | |
| CI-GCL2026.04 | 73.85 | — | |
| WL kernelCategory=Graph Kernel Methods, Validation=Tenfold cross-validation2024.12 | 73.8 | — | |
| lrGAE 7Approach Category=Structure-based2024.10 | 73.8 | — | |
| lrGAE 8Approach Category=Structure-based2024.10 | 73.8 | — | |
| MSPN2026.05 | 73.7 | — | |
| GCL-SPAN2026.04 | 73.65 | — | |
| S2GAEApproach Category=Structure-based2024.10 | 73.6 | — | |
| GMT2026.03 | 73.5 | — | |
| CWN2026.05 | 73.5 | — | |
| WLHN2026.05 | 73.4 | — | |
| GSN2026.05 | 73.36 | — | |
| Ring-GNN (w/ degree)Input node features=node degrees (integer)2019.05 | 73.3 | 4.9 | |
| ECPLandmarks (K)=200, Evaluation Protocol=5 seeds x 10 folds2026.05 | 73.3 | — | |
| AutoGCLLearning Paradigm=Unsupervised2024.08 | 73.3 | — | |
| GCN2026.05 | 73.3 | — | |
| BGRLApproach Category=Standard GCL2024.10 | 73.2 | — | |
| CGSValidation=10-fold cross-validation, Setting=Inductive2022.10 | 73.1 | 3.3 | |
| sGNN2Input node features=identical features2019.05 | 73.1 | 5.2 | |
| GTR2026.05 | 73.1 | — | |
| TopNets2026.05 | 73.1 | — | |
| InfoGraph2026.04 | 73.03 | — | |
| InfographLearning Paradigm=Unsupervised2024.08 | 73.03 | — | |
| Ring-GNNInput node features=identical features2019.05 | 73 | 5.4 | |
| PPGNsValidation=Tenfold cross-validation2024.12 | 73 | — | |
| PPGNs2026.05 | 73 | — | |
| GRACEApproach Category=Standard GCL2024.10 | 73 | — | |
| PPGNPeak GPU Memory (MB)=21932026.05 | 73 | — | |
| GraphTrail2026.03 | 72.9 | — | |
| sGNN5Input node features=identical features2019.05 | 72.8 | 3.8 | |
| ASAP2026.03 | 72.8 | — | |
| sGNN1Input node features=identical features2019.05 | 72.7 | 4.9 | |
| MINCUTPOOL2026.03 | 72.7 | — | |
| WITTOPOPOOL2026.03 | 72.6 | — | |
| PathNN2026.05 | 72.6 | — | |
| DiffPoolLearning Paradigm=Supervised2024.08 | 72.6 | — | |
| SeeGeraApproach Category=Feature-based2024.10 | 72.6 | — |