Graph Classification on IMDB-B (Accuracy, Avg. Rank)
74.3AccuracyG-TUNING
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| G-TUNINGpre-training=standard2023.12 | 74.3 | 1.1 | |
| BSS2023.12 | 73.2 | 4.7 | |
| InfoGraphLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 73.03 | 7 | |
| Feature Map2023.12 | 72.9 | 5.6 | |
| VGAE-Tuning2023.12 | 72.5 | 6.6 | |
| GTOT-Tuning2023.12 | 72.3 | 6 | |
| MGCLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 72.28 | 1.42 | |
| StochNorm2023.12 | 72.2 | 6 | |
| Vanilla Tuning2023.12 | 72.1 | 6.6 | |
| G-TUNINGpre-training=none2023.12 | 72.1 | 6.7 | |
| AutoGCLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 72 | 7.14 | |
| DRGCLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 72 | 3.57 | |
| L2_SP2023.12 | 71.9 | 6 | |
| RGCLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 71.9 | 4.71 | |
| DELTA2023.12 | 71.8 | 6.6 | |
| simGRACELearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 71.3 | 4.14 | |
| GraphCLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 71.14 | 6.42 | |
| JOAOv2Learning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 70.83 | 7.28 | |
| JOAOLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 70.21 | 7.85 | |
| AD-GCLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 70.2 | 7.28 | |
| HTG-GCLLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 69.1 | 1 | |
| RGCLLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68.9 | 3.4 | |
| SimGRACELearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68.6 | 4.6 | |
| InfoGraphLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68.5 | 8.2 | |
| CellCLATLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68.5 | 3 | |
| HTMLLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68.4 | 7.8 | |
| JOAOv2Learning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68.3 | 6.8 | |
| DRGCLLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 68 | 6.4 | |
| GraphCLLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 67.8 | 6 | |
| ADGCLLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 67.7 | 10.8 | |
| JOAOLearning Protocol=Semi-supervised, Labels=10%, Evaluation=10-fold CV with SVM2025.11 | 67.6 | 6.6 | |
| MVGRLLearning Protocol=unsupervised representation learning, Evaluation Protocol=linear evaluation protocol, Cross-Validation=10-fold2025.10 | 63.6 | 8.85 | |
| Supervised2023.12 | 52.4 | 10.1 |