Node classification on Computers (test)
92.97Mean AccuracyDirGNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DirGNN2024.10 | 92.97 | — | |
| CoEDedge direction learning=false2024.10 | 92.88 | — | |
| Co-GNN2024.10 | 92.76 | — | |
| TabPFN-GN2025.12 | 92.71 | — | |
| Chung2024.10 | 92.57 | — | |
| D2MoEBackbone=GCN, Method Category=Ours2026.04 | 92.23 | — | |
| D2MoEBackbone=GAT, Method Category=Ours2026.04 | 92.14 | — | |
| D2MoEBackbone=SAGE, Method Category=Ours2026.04 | 92.01 | — | |
| MOWST-2024Method Category=GRAPH MOE2026.04 | 92 | — | |
| NODEMOE-2024Method Category=GRAPH MOE2026.04 | 91.87 | — | |
| GGCN2024.10 | 91.81 | — | |
| GPRGNN-2021Method Category=HETERO2026.04 | 91.8 | — | |
| SignGT2023.10 | 91.71 | — | |
| H2GCN-2020Method Category=HETERO2026.04 | 91.69 | — | |
| ACMGCN-2022Method Category=HETERO2026.04 | 91.66 | — | |
| MOSCAT-2025Method Category=GRAPH MOE2026.04 | 91.65 | — | |
| DAMOE-2025Method Category=GRAPH MOE2026.04 | 91.57 | — | |
| DIFFORMER-2023Method Category=GT2026.04 | 91.52 | — | |
| EXPHORMER-2023Method Category=GT2026.04 | 91.46 | — | |
| GATMethod Category=VANILLA2026.04 | 91.44 | — | |
| GMOE-2023Method Category=GRAPH MOE2026.04 | 91.37 | — | |
| SAGE2024.10 | 91.2 | — | |
| GraphSAGE2025.12 | 91.2 | — | |
| GraphGPS2025.12 | 91.19 | — | |
| GCNMethod Category=VANILLA2026.04 | 91.17 | — | |
| NodeFormer2023.10 | 91.12 | — | |
| SPGCLTraining Data=X, A2026.06 | 91.04 | — | |
| FSGNN-2022Method Category=HETERO2026.04 | 91.03 | — | |
| Specformer2023.10 | 91.02 | — | |
| SAGEMethod Category=VANILLA2026.04 | 90.94 | — | |
| SIGNATraining Data=X, A2026.06 | 90.9 | — | |
| FLODE2024.10 | 90.88 | — | |
| GAT2023.10 | 90.78 | — | |
| GAT2024.10 | 90.78 | — | |
| GAT2025.12 | 90.78 | — | |
| HiGCNOrder=22023.09 | 90.76 | — | |
| SGFORMER-2024Method Category=GT2026.04 | 90.7 | — | |
| CSGCLTraining Data=X, A2026.06 | 90.67 | — | |
| HiGCNOrder=32023.09 | 90.65 | — | |
| All featuresselection_ratio=100%, backbone=GNN2025.10 | 90.58 | — | |
| Sheaf2024.10 | 90.56 | — | |
| HiGCNOrder=12023.09 | 90.5 | — | |
| HiGCNOrder=42023.09 | 90.35 | — | |
| MIselection_ratio=5%, backbone=GNN2025.10 | 90.33 | — | |
| StrGCLTraining Data=X, A2026.06 | 90.32 | — | |
| MagNet2024.10 | 90.3 | — | |
| SGC2023.10 | 90.22 | — | |
| NAGFORMER-2023Method Category=GT2026.04 | 90.22 | — | |
| NPTselection_ratio=5%, backbone=GNN2025.10 | 90.09 | — | |
| TFIselection_ratio=5%, backbone=GNN2025.10 | 90.09 | — | |
| hattrselection_ratio=5%, backbone=GNN2025.10 | 90.09 | — | |
| SGRLTraining Data=X, A2026.06 | 90.03 | — | |
| ANS-GT-2022Method Category=GT2026.04 | 90.01 | — | |
| GCATraining Data=X, A2026.06 | 89.95 | — | |
| NPT-maskselection_ratio=5%, backbone=GNN2025.10 | 89.94 | — | |
| AFGRL*Training Data=X, A2026.06 | 89.88 | — | |
| GCN2024.10 | 89.65 | — | |
| GCN2025.12 | 89.65 | — | |
| Space OptimumMaximum refinement budget=1002025.07 | 89.59 | — | |
| LocalGCLTraining Data=X, A2026.06 | 89.59 | — | |
| FAGCN-2021Method Category=HETERO2026.04 | 89.54 | — | |
| GCN2023.10 | 89.53 | — | |
| SCETraining Data=X, A2026.06 | 89.45 | — | |
| BGRLTraining Data=X, A2026.06 | 89.41 | — | |
| GPRGNN2023.10 | 89.32 | — | |
| hGEselection_ratio=5%, backbone=GNN2025.10 | 89.24 | — | |
| M-DESIGNMaximum refinement budget=1002025.07 | 89.22 | — | |
| hEucselection_ratio=5%, backbone=GNN2025.10 | 89.19 | — | |
| GloGNN2023.10 | 89.12 | — | |
| E2NegTraining Data=X, A2026.06 | 89.02 | — | |
| GRACETraining Data=X, A2026.06 | 88.78 | — | |
| DesiGNNMaximum refinement budget=1002025.07 | 88.4 | — | |
| EAMaximum refinement budget=1002025.07 | 88.28 | — | |
| Rnd.selection_ratio=5%, backbone=GNN2025.10 | 88.27 | — | |
| RandomMaximum refinement budget=1002025.07 | 88.25 | — | |
| RLMaximum refinement budget=1002025.07 | 88.25 | — | |
| ProGCLTraining Data=X, A2026.06 | 88.09 | — | |
| GraphNASMaximum refinement budget=1002025.07 | 87.94 | — | |
| DGITraining Data=X, A2026.06 | 87.89 | — | |
| AutoTransferMaximum refinement budget=1002025.07 | 87.72 | — | |
| BernNet2023.09 | 87.64 | — | |
| Auto-GNNMaximum refinement budget=1002025.07 | 87.59 | — | |
| ChebNet2023.09 | 87.54 | — | |
| KBGMaximum refinement budget=1002025.07 | 87.38 | — | |
| BernNettest subset=observed2022.12 | 87.17 | — | |
| SGCModel Type=Non-Deep2022.02 | 87.14 | — | |
| FAGCN2023.10 | 87.11 | — | |
| NFGNNtest subset=observed2022.12 | 86.91 | — | |
| GPRGNN2023.09 | 86.85 | — | |
| ASGCModel Type=Non-Deep2022.02 | 86.72 | — | |
| GCNTraining Data=X, A, Y2026.06 | 86.51 | — | |
| APPNPtest subset=observed2022.12 | 86.3 | — | |
| GPRGNNtest subset=observed2022.12 | 86.09 | — | |
| APPNP2023.09 | 85.32 | — | |
| FGN2026.06 | 85.1 | — | |
| MLPMethod Category=VANILLA2026.04 | 85.01 | — | |
| MAGCN2026.06 | 84.7 | — | |
| GInterNet2026.06 | 84.5 | — | |
| VANILLA GTMethod Category=GT2026.04 | 84.41 | — | |
| GCN-IED2026.06 | 84.2 | — |