Graph Retrieval on MR (test)
90.2MAPIsoNet++
Evaluation results
| Method | Links | ||
|---|---|---|---|
| IsoNet++Aggregation strategy=multi-round, Graph representation level=Edge2025.10 | 90.2 | 80.3 | |
| IsoNet++Aggregation strategy=multi-layer, Graph representation level=Edge2025.10 | 88.6 | 77.6 | |
| IsoNet++Aggregation strategy=multi-round, Graph representation level=Node2025.10 | 87.4 | 75.5 | |
| IsoNet++Aggregation strategy=multi-layer, Graph representation level=Node2025.10 | 84.1 | 70.7 | |
| multi-round IsoNet++ (Edge)Training Dataset=AIDS2025.10 | 83.6 | — | |
| IsoNetGraph representation level=Edge2025.10 | 77.4 | 60.1 | |
| IsoNetGraph representation level=Node2025.10 | 73.8 | 53.2 | |
| GMN2025.10 | 70.8 | 49 | |
| multi-round IsoNet++ (Node)Training Dataset=Mutag2025.10 | 65.5 | — | |
| GEN2025.10 | 63.1 | 39.1 | |
| Neuromatch2025.10 | 60.6 | 38.5 | |
| EGSC2025.10 | 58 | 32.5 | |
| GREED2025.10 | 52.8 | 27.7 | |
| GraphSim2025.10 | 45.3 | 23 | |
| SimGNN2025.10 | 37.9 | 16 | |
| GOTSim2025.10 | 31.7 | 11.6 | |
| H2MN2025.10 | 24.3 | 6.9 |