Node Classification On Non Homophilic 12
Metrics
1:1 Accuracy
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | 1:1 Accuracy |
|---|---|
| revisiting-heterophily-for-graph-neural | 55.19 ± 1.49 |
| large-scale-learning-on-non-homophilous | 61.81 ± 1.80 |
| deformable-graph-convolutional-networks | 62.56 ± 1.31 |
| revisiting-heterophily-for-graph-neural | 67.06 ± 1.66 |
| addressing-heterophily-in-node-classification | 73.56 ± 1.62 |
| scale-invariance-of-graph-neural-networks | 76.0±2.0 |
| neural-sheaf-diffusion-a-topological | 53.17 ± 1.31 |
| two-sides-of-the-same-coin-heterophily-and | 55.17 ± 1.58 |
| breaking-the-limit-of-graph-neural-networks | 48.85 ± 0.78 |
| finding-global-homophily-in-graph-neural | 57.88 ± 1.76 |
| non-local-graph-neural-networks | 33.7 ± 1.5 |
| revisiting-heterophily-for-graph-neural | 51.8 ± 1.5 |
| joint-adaptive-feature-smoothing-and-topology | 46.31 ± 2.46 |
| simple-and-deep-graph-convolutional-networks-1 | 38.47 ± 1.58 |
| neural-sheaf-diffusion-a-topological | 56.34 ± 1.32 |
| revisiting-heterophily-for-graph-neural | 66.98 ± 1.71 |
| non-local-graph-neural-networks | 59.0 ± 1.2 |
| finding-global-homophily-in-graph-neural | 57.54 ± 1.39 |
| edge-directionality-improves-learning-on | 75.31±1.92 |
| revisiting-heterophily-for-graph-neural | 40.02 ± 0.96 |
| revisiting-heterophily-for-graph-neural | 67.07 ± 1.65 |
| generalizing-graph-neural-networks-beyond | 36.48 ± 1.86 |
| geom-gcn-geometric-graph-convolutional-1 | 38.15 ± 0.92 |
| revisiting-heterophily-for-graph-neural | 45.00 ± 1.4 |
| revisiting-heterophily-for-graph-neural | 67.4 ± 2.21 |
| beyond-low-frequency-information-in-graph | 30.83 ± 0.69 |
| non-local-graph-neural-networks | 56.8 ± 2.5 |
| mixhop-higher-order-graph-convolution | 43.80 ± 1.48 |
| neural-sheaf-diffusion-a-topological | 54.78 ± 1.81 |