| Edge2Node: Reducing Edge Prediction to Node Classification | | 0.9515 ± 0.1410 | No | 0.9546 ± 0.1270 | 526851 | E2N | 2023-11-06 |
| []() | | 0.7129 ± 0.0018 | No | 0.7385 ± 0.0099 | 1064446212 | HyperFusion | |
| GIDN: A Lightweight Graph Inception Diffusion Network for High-efficient Link Prediction | | 0.7096 ± 0.0055 | No | 0.9620 ± 0.0040 | 60449025 | GIDN@YITU | 2022-10-04 |
| []() | | 0.7087 ± 0.0033 | No | 1.0000 ± 0.0000 | 34980864 | PLNLP + SIGN | |
| Pairwise Learning for Neural Link Prediction | ✓ Link | 0.7059 ± 0.0029 | No | 1.0000 ± 0.0000 | 34980864 | PLNLP (random walk aug.) | 2021-12-06 |
| []() | | 0.7012 ± 0.0016 | No | 1.0000 ± 0.0000 | 30191104 | HOP-REC | |
| Global Attention Improves Graph Networks Generalization | ✓ Link | 0.6909 ± 0.0055 | No | 1.0000 ± 0.0000 | 35200656 | PLNLP+ LRGA | 2020-06-14 |
| Pairwise Learning for Neural Link Prediction | ✓ Link | 0.6872 ± 0.0052 | No | 1.0000 ± 0.0000 | 35112192 | PLNLP (val as input) | 2021-12-06 |
| Reconsidering the Performance of GAE in Link Prediction | ✓ Link | 0.6816 ± 0.0041 | No | 1.0000 ± 0.0000 | 126669825 | Refined-GAE | 2024-11-06 |
| []() | | 0.6792 ± 0.0074 | No | 0.6771 ± 0.0083 | 483363845 | TopoLink | |
| Simplifying Subgraph Representation Learning for Scalable Link Prediction | ✓ Link | 0.6683 ± 0.0030 | No | 0.9861 ± 0.0006 | 5913025 | S3GRL (PoS Plus) | 2023-01-29 |
| []() | | 0.6636 ± 0.5876 | No | 0.6631 ± 0.0021 | 3284065 | ELPH | |
| []() | | 0.6572 ± 0.0053 | No | 0.6621 ± 0.0016 | 1184867 | BUDDY | |
| Edge Proposal Sets for Link Prediction | ✓ Link | 0.6548 ± 0.0000 | No | 0.9735 ± 0.0000 | 0 | Adamic Adar+Edge Proposal Set | 2021-06-30 |
| Labeling Trick: A Theory of Using Graph Neural Networks for Multi-Node Representation Learning | ✓ Link | 0.6474 ± 0.0043 | No | 0.6495 ± 0.0043 | 501570 | SEAL-nofeat (val as input) | 2020-10-30 |
| []() | | 0.6417 ± 0.0000 | No | 0.6349 ± 0.0000 | 0 | Adamic Adar | |
| []() | | 0.6137 ± 0.0000 | No | 0.6036 ± 0.0000 | 0 | Common Neighbor | |
| Labeling Trick: A Theory of Using Graph Neural Networks for Multi-Node Representation Learning | ✓ Link | 0.5471 ± 0.0049 | No | 0.6495 ± 0.0043 | 501570 | SEAL-nofeat | 2020-10-30 |
| Inductive Representation Learning on Large Graphs | ✓ Link | 0.5463 ± 0.0112 | No | 0.5688 ± 0.0077 | 460289 | GraphSAGE (val as input) | 2017-06-07 |
| Network In Graph Neural Network | | 0.5359 ± 0.0056 | No | 0.6281 ± 0.0046 | 591873 | NGNN + GraphSAGE | 2021-11-23 |
| Network In Graph Neural Network | | 0.5348 ± 0.0040 | No | 0.6273 ± 0.0040 | 428033 | NGNN + GCN | 2021-11-23 |
| DeeperGCN: All You Need to Train Deeper GCNs | ✓ Link | 0.5273 ± 0.0047 | No | 0.6187 ± 0.0045 | 117383 | DeeperGCN | 2020-06-13 |
| Global Attention Improves Graph Networks Generalization | ✓ Link | 0.5221 ± 0.0072 | No | 0.6088 ± 0.0059 | 1069489 | LRGA + GCN | 2020-06-14 |
| On the effect of the average clustering coefficient on topology-based link prediction in featureless graphs | ✓ Link | 0.5050 ± 0.0000 | No | 0.6098 ± 0.0000 | 0 | Jaccard Index | 2025-01-12 |
| DeepWalk: Online Learning of Social Representations | ✓ Link | 0.5037 ± 0.0034 | No | Please tell us | 61390187 | DeepWalk | 2014-03-26 |
| node2vec: Scalable Feature Learning for Networks | ✓ Link | 0.4888 ± 0.0054 | No | 0.5703 ± 0.0052 | 30322945 | Node2vec | 2016-07-03 |
| Inductive Representation Learning on Large Graphs | ✓ Link | 0.4810 ± 0.0081 | No | 0.5688 ± 0.0077 | 460289 | GraphSAGE | 2017-06-07 |
| Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 0.4714 ± 0.0145 | No | 0.5263 ± 0.0115 | 296449 | GCN (val as input) | 2016-09-09 |
| VQ-GNN: A Universal Framework to Scale up Graph Neural Networks using Vector Quantization | ✓ Link | 0.4673 ± 0.0164 . | | | | VQ-GNN (SAGE-Mean) | 2021-10-27 |
| Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 0.4475 ± 0.0107 | No | 0.5263 ± 0.0115 | 296449 | GCN | 2016-09-09 |
| VQ-GNN: A Universal Framework to Scale up Graph Neural Networks using Vector Quantization | ✓ Link | 0.4316 ± 0.0134 | | | | VQ-GNN (GCN) | 2021-10-27 |
| VQ-GNN: A Universal Framework to Scale up Graph Neural Networks using Vector Quantization | ✓ Link | 0.4102 ± 0.0099 | | | | VQ-GNN (GAT) | 2021-10-27 |
| Open Graph Benchmark: Datasets for Machine Learning on Graphs | ✓ Link | 0.3886 ± 0.0029 | No | 0.4896 ± 0.0029 | 60514049 | Matrix Factorization | 2020-05-02 |
| Inductive Representation Learning on Large Graphs | ✓ Link | | | | 460289 | GraphSAGE (val as input) | 2017-06-07 |