Node Classification on Cora (60/20/20 random split)
89.89AccuracyP2GNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| P2GNNBackbone=ACM-GCN2026.03 | 89.89 | — | |
| JacobiConv2025.12 | 89.61 | — | |
| ACM-Snowball-32021.09 | 89.59 | — | |
| GAT+JK2021.09 | 89.52 | — | |
| GAT+JK2026.03 | 89.52 | — | |
| ACMII-Snowball-32021.09 | 89.36 | — | |
| Snowball-32021.09 | 89.33 | — | |
| Snowball-32026.03 | 89.33 | — | |
| ATLAS-LPF-NF2025.12 | 89.2 | — | |
| ACM-GCNII2021.09 | 89.1 | — | |
| ATLAS-NF2025.12 | 89.08 | — | |
| ACM-GCNII*2021.09 | 89 | — | |
| ACMII-GCN2021.09 | 89 | — | |
| GCNII2021.09 | 88.98 | — | |
| GCNII2026.03 | 88.98 | — | |
| ACMII-Snowball-22021.09 | 88.95 | — | |
| GCNII*2021.09 | 88.93 | — | |
| ClenshawGCN2022.10 | 88.9 | — | |
| ClenshawGCNSplits=20 random 60%/20%/20% train/validation/test2022.10 | 88.9 | — | |
| FAGCN2025.12 | 88.85 | — | |
| FAGCN2021.09 | 88.85 | — | |
| FAGCN2026.03 | 88.85 | — | |
| ACM-Snowball-22021.09 | 88.83 | — | |
| ChebNetIISplits=20 random 60%/20%/20% train/validation/test2022.10 | 88.71 | — | |
| Snowball-22021.09 | 88.64 | — | |
| Snowball-22026.03 | 88.64 | — | |
| ACM-GCN2021.09 | 88.62 | — | |
| GGCN2026.03 | 88.57 | — | |
| ACM-GCN2026.03 | 88.56 | — | |
| GPRGNNSplits=20 random 60%/20%/20% train/validation/test2022.10 | 88.54 | — | |
| BernNet2026.03 | 88.52 | — | |
| BernNetSplits=20 random 60%/20%/20% train/validation/test2022.10 | 88.51 | — | |
| GCNII2022.10 | 88.46 | — | |
| ATLAS-LPF2025.12 | 88.46 | — | |
| OrderedGNN2025.12 | 88.37 | — | |
| ATLAS2025.12 | 88.37 | — | |
| GloGNN2025.12 | 88.31 | — | |
| VR-GNN2026.03 | 88.27 | — | |
| APPNPSplits=20 random 60%/20%/20% train/validation/test2022.10 | 88.16 | — | |
| M2M-GNN2025.12 | 88.12 | — | |
| tGNN2022.05 | 88.08 | — | |
| ACM-GCN2025.12 | 87.91 | — | |
| GCN2022.05 | 87.78 | — | |
| GCN2021.09 | 87.78 | — | |
| GAT2025.12 | 87.74 | — | |
| ACM-SGC-22021.09 | 87.64 | — | |
| H2GCN2022.05 | 87.52 | — | |
| H2GCN2022.10 | 87.52 | — | |
| H2GCN2025.12 | 87.52 | — | |
| H2GCN2021.09 | 87.52 | — | |
| H2GCN2026.03 | 87.52 | — | |
| FSGNN2025.12 | 87.51 | — | |
| SAGE2025.12 | 87.5 | — | |
| ChebNetSplits=20 random 60%/20%/20% train/validation/test2022.10 | 87.32 | — | |
| GCN2022.10 | 87.18 | — | |
| ARMASplits=20 random 60%/20%/20% train/validation/test2022.10 | 87.13 | — | |
| GBK-GNN2025.12 | 87.09 | — | |
| GCN2025.12 | 87.01 | — | |
| GCN+JK2022.10 | 86.9 | — | |
| GCN+JK2021.09 | 86.9 | — | |
| GCN+JK2026.03 | 86.9 | — | |
| GAT2022.05 | 86.86 | — | |
| ACM-SGC-12021.09 | 86.63 | — | |
| GraphSAGE2022.05 | 86.58 | — | |
| CMGNN2025.12 | 85.76 | — | |
| SGC-22021.09 | 85.48 | — | |
| SGC-22026.03 | 85.48 | — | |
| Geom-GCN2023.12 | 85.27 | — | |
| Geom-GCN†2021.09 | 85.27 | — | |
| SGC-12021.09 | 85.12 | — | |
| UGCN2023.12 | 84 | — | |
| SimP-GCN2023.12 | 82.8 | — | |
| LinkX2025.12 | 82.62 | — | |
| GPRGNN2022.05 | 79.51 | — | |
| GPR-GNN2025.12 | 79.51 | — | |
| GPRGNN*2021.09 | 79.51 | — | |
| GPRGNN2026.03 | 79.51 | — | |
| APPNP2022.05 | 79.41 | — | |
| APPNP*2021.09 | 79.41 | — | |
| GraphSAGE2021.09 | 78.24 | — | |
| MLP2022.10 | 76.89 | — | |
| GAT*2021.09 | 76.7 | — | |
| GAT2026.03 | 76.7 | — | |
| MLP-2*2021.09 | 76.44 | — | |
| MLP2026.03 | 76.44 | — | |
| MLP2025.12 | 75.44 | — | |
| GloGNN++2026.03 | 74.87 | — | |
| CoLinkDistsplit=60%/20%/20%2023.06 | 70.32 | — | |
| MixHop2022.05 | 65.65 | — | |
| MixHop2022.10 | 65.65 | — | |
| MixHop2021.09 | 65.65 | — | |
| Best C&S modelParameter Δ=-98.37%, Time=0.5 s2020.10 | — | 1.09 |