| GNNDLD: Graph Neural Network with Directional Label Distribution | | 91.95±0.19 | GNNDLD | 2024-02-26 |
| Neighborhood Homophily-Guided Graph Convolutional Network | ✓ Link | 91.56 ± 0.50 | NHGCN | 2023-10-21 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 91.44 ± 0.59 | ACM-Snowball-3 | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 91.31 ± 0.6 | ACMII-Snowball-3 | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.96 ± 0.62 | ACMII-GCN+ | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.81 ± 0.52 | ACM-Snowball-2 | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.74 ± 0.5 | ACMII-GCN | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.66 ± 0.47 | ACM-GCN | 2022-10-14 |
| The Split Matters: Flat Minima Methods for Improving the Performance of GNNs | ✓ Link | 90.64 ± 0.46% | Graph-MLP + SAF | 2023-06-15 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.63 ± 0.56 | ACMII-GCN++ | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.56 ± 0.39 | ACMII-Snowball-2 | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.46 ± 0.69 | ACM-GCN+ | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.39 ± 0.33 | ACM-GCN++ | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.18 ± 0.51 | ACM-GCNII* | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.12 ± 0.4 | ACM-GCNII | 2022-10-14 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 90.09 ± 0.68 | GCN+JK | 2022-10-14 |
| Geom-GCN: Geometric Graph Convolutional Networks | ✓ Link | 90.05 | Geom-GCN* | 2020-02-13 |
| Beyond Low-frequency Information in Graph Convolutional Networks | ✓ Link | 89.98 ± 0.54 | FAGCN | 2021-01-04 |
| Simple and Deep Graph Convolutional Networks | ✓ Link | 89.98 ± 0.52 | GCNII* | 2020-07-04 |
| Node-oriented Spectral Filtering for Graph Neural Networks | ✓ Link | 89.89±0.68 | NFGNN | 2022-12-07 |
| Simple and Deep Graph Convolutional Networks | ✓ Link | 89.8 ± 0.3 | GCNII | 2020-07-04 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 89.15 ± 0.87 | GAT+JK | 2022-10-14 |
| Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks | ✓ Link | 89.04 ± 0.49 | Snowball-2 | 2019-06-05 |
| Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 88.9 ± 0.32 | GCN | 2016-09-09 |
| Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks | ✓ Link | 88.8 ± 0.82 | Snowball-3 | 2019-06-05 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 88.79 ± 0.5 | ACM-SGC-2 | 2022-10-14 |
| BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation | ✓ Link | 88.48 ± 0.41 | BernNet | 2021-06-21 |
| Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs | ✓ Link | 87.78 ± 0.28 | H2GCN | 2020-06-20 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 87.75 ± 0.88 | ACM-SGC-1 | 2022-10-14 |
| MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing | ✓ Link | 87.04 ± 4.10 | MixHop | 2019-04-30 |
| Inductive Representation Learning on Large Graphs | ✓ Link | 86.85 ± 0.11 | GraphSAGE | 2017-06-07 |
| Revisiting Heterophily For Graph Neural Networks | ✓ Link | 86.43 ± 0.13 | MLP-2 | 2022-10-14 |
| Simplifying Graph Convolutional Networks | ✓ Link | 85.5 ± 0.76 | SGC-1 | 2019-02-19 |
| Simplifying Graph Convolutional Networks | ✓ Link | 85.36 ± 0.52 | SGC-2 | 2019-02-19 |
| Adaptive Universal Generalized PageRank Graph Neural Network | ✓ Link | 85.07 ± 0.09 | GPRGNN | 2020-06-14 |
| Predict then Propagate: Graph Neural Networks meet Personalized PageRank | ✓ Link | 85.02 ± 0.09 | APPNP | 2018-10-14 |
| Graph Attention Networks | ✓ Link | 83.28 ± 0.12 | GAT | 2017-10-30 |