Graph Classification on IMDB-M
79.4AccuracyGTR
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GTR2026.05 | 79.4 | — | |
| AutoGELType=AutoGNN2021.12 | 56.8 | — | |
| TL-GNNValidation=Tenfold cross-validation2024.12 | 55.1 | — | |
| CCIN2026.05 | 54.7 | — | |
| DiGGRLearning Paradigm=Unsupervised2024.08 | 54.69 | — | |
| SNNValidation=Tenfold cross-validation, Variant=SNN(α, (1, 1)), γ=0.52024.12 | 54.53 | — | |
| (s)GIN-0neighborhood enlargement=true, epsilon=02019.05 | 54.52 | — | |
| GSNValidation=Tenfold cross-validation2024.12 | 54.3 | — | |
| GSNEvaluation Protocol=Standard eval.2026.05 | 54.3 | — | |
| VEPMLearning Paradigm=Supervised/Semi-supervised2024.08 | 54.1 | — | |
| MaxCut-Pool2026.05 | 54.1 | — | |
| G2N2Peak GPU Memory (MB)=4002026.05 | 54 | — | |
| (s)GIN-epsilonneighborhood enlargement=true2019.05 | 53.62 | — | |
| UGformerArchitecture=Variant 1, Inference Setting=Inductive, Evaluation Protocol=10-fold cross-validation2019.09 | 53.6 | — | |
| UGT# Parameters=48,6982023.11 | 53.6 | — | |
| UGT2023.11 | 53.6 | — | |
| MGNNIValidation=10-fold cross-validation, Setting=Inductive2022.10 | 53.5 | 2.8 | |
| SEG-BERT*Strategy=Best results obtained with all three unification strategies2020.02 | 53.4 | — | |
| CH-CL2026.04 | 53.2 | — | |
| GCBM-E2026.03 | 53.1 | — | |
| ShareGNNEvaluation Protocol=Standard eval.2026.05 | 53.1 | — | |
| AIA2022.11 | 53.02 | — | |
| GraphMAE2 + COREmasking process=GraphMAE22025.12 | 53 | — | |
| WITTOPOPOOL2026.03 | 52.9 | — | |
| GRDL2026.03 | 52.9 | — | |
| SAGEConvPeak GPU Memory (MB)=182026.05 | 52.9 | — | |
| M-Mixup2022.11 | 52.89 | — | |
| TopoGCL2026.04 | 52.81 | — | |
| GNTK2019.05 | 52.8 | — | |
| GraphMAE + COREmasking process=GraphMAE2025.12 | 52.8 | — | |
| GNTKCategory=Graph Kernel Methods, Validation=Tenfold cross-validation2024.12 | 52.8 | — | |
| Mincut-Pool2026.05 | 52.8 | — | |
| fCWNPeak GPU Memory (MB)=192026.05 | 52.8 | — | |
| GraphMAE22025.12 | 52.7 | — | |
| CINValidation=Tenfold cross-validation2024.12 | 52.7 | — | |
| CINEvaluation Protocol=Standard eval.2026.05 | 52.7 | — | |
| GraphMAE2Approach Category=Feature-based2024.10 | 52.7 | — | |
| lrGAE 8Approach Category=Structure-based2024.10 | 52.7 | — | |
| CINPeak GPU Memory (MB)=5172026.05 | 52.7 | — | |
| CAL2022.11 | 52.6 | — | |
| WWL2026.03 | 52.6 | — | |
| MaskGAEApproach Category=Structure-based2024.10 | 52.6 | — | |
| DDMProtocol=Unsupervised representation learning2023.06 | 52.53 | — | |
| GCMAE2025.12 | 52.5 | — | |
| GCBM2026.03 | 52.5 | — | |
| SINValidation=Tenfold cross-validation2024.12 | 52.5 | — | |
| DisenSemiLearning Paradigm=Supervised/Semi-supervised2024.08 | 52.5 | — | |
| HMH2026.05 | 52.5 | — | |
| GAEfApproach Category=Feature-based2024.10 | 52.5 | — | |
| S2GAEApproach Category=Structure-based2024.10 | 52.5 | — | |
| StructPoolevaluation_protocol=10-fold cross-validation2021.07 | 52.47 | — | |
| (s)GCNneighborhood enlargement=true2019.05 | 52.4 | — | |
| SINEvaluation Protocol=Standard eval.2026.05 | 52.4 | — | |
| GCNPeak GPU Memory (MB)=182026.05 | 52.4 | — | |
| GIN-0neighborhood enlargement=false, epsilon=02019.05 | 52.3 | — | |
| GIN2019.05 | 52.3 | — | |
| GIN-0Learning paradigm=Supervised Approaches2021.10 | 52.3 | — | |
| GINType=Manual GNNs2021.12 | 52.3 | — | |
| GIN-0Inference Setting=Inductive, Evaluation Protocol=10-fold cross-validation2019.09 | 52.3 | — | |
| GINValidation=10-fold cross-validation, Setting=Inductive2022.10 | 52.3 | 2.8 | |
| GINInput node features=node degrees (one-hot)2019.05 | 52.3 | 2.8 | |
| GINProtocol=Unsupervised representation learning2023.06 | 52.3 | — | |
| GIN2025.12 | 52.3 | — | |
| GINValidation=Tenfold cross-validation2024.12 | 52.3 | — | |
| GINConfig Category=Top NT, Landmarks (K)=200, Evaluation Protocol=5 seeds x 10 folds2026.05 | 52.3 | — | |
| GINLearning Paradigm=Supervised2024.08 | 52.3 | — | |
| GIN-02026.05 | 52.3 | — | |
| HGNNPeak GPU Memory (MB)=182026.05 | 52.3 | — | |
| sCWNcPeak GPU Memory (MB)=192026.05 | 52.3 | — | |
| SEG-BERTStrategy=padding/pruning, Graph residual terms=none2020.02 | 52.27 | — | |
| RADE-OFSBackbone=GCN2026.05 | 52.24 | — | |
| BGRLApproach Category=Standard GCL2024.10 | 52.2 | — | |
| AUG-MAEApproach Category=Feature-based2024.10 | 52.2 | — | |
| lrGAE 6Approach Category=Structure-based2024.10 | 52.2 | — | |
| lrGAE 7Approach Category=Structure-based2024.10 | 52.2 | — | |
| DropNodeBackbone=GCN2026.05 | 52.2 | — | |
| FLAG2022.11 | 52.16 | — | |
| GCL-SPAN2026.04 | 52.16 | — | |
| GIN-epsilonneighborhood enlargement=false2019.05 | 52.1 | — | |
| GIN-eLearning paradigm=Supervised Approaches2021.10 | 52.1 | — | |
| EIGNNValidation=10-fold cross-validation, Setting=Inductive2022.10 | 52.1 | 2.9 | |
| OT-GNN2026.03 | 52.1 | — | |
| MSPN2026.05 | 52.1 | — | |
| GraphMAEApproach Category=Feature-based2024.10 | 52.1 | — | |
| RADE-OFBackbone=GCN2026.05 | 52.09 | — | |
| DropoutBackbone=GCN2026.05 | 52.06 | — | |
| GIN2026.05 | 52 | — | |
| EPAGCLApproach Category=Standard GCL2024.10 | 52 | — | |
| GraphGPS2026.05 | 51.93 | — | |
| GCNBackbone=GCN2026.05 | 51.93 | — | |
| GCNneighborhood enlargement=false2019.05 | 51.9 | — | |
| GCNLearning paradigm=Supervised Approaches2021.10 | 51.9 | — | |
| GCNType=Manual GNNs2021.12 | 51.9 | — | |
| GCNInference Setting=Inductive, Evaluation Protocol=10-fold cross-validation2019.09 | 51.9 | — | |
| GCN2023.11 | 51.9 | — | |
| GCN2023.11 | 51.9 | — | |
| DGCLLearning Paradigm=Unsupervised2024.08 | 51.9 | — | |
| GFN2019.05 | 51.8 | — | |
| GFNInference Setting=Inductive, Evaluation Protocol=10-fold cross-validation2019.09 | 51.8 | — | |
| AUG-MAE2025.12 | 51.8 | — |