Check leaderboards for
The bold method name indicates that the implementation is official (by the author of the original paper).
Package denotes the required package version for each dataset to be eligible for the leaderboard.
Note that the evaluation metric computation has been updated in versioin 0.8.0, please refer to the examples/nodeproppred/ folder to see corrected metric computations.
For previous version of leaderboards, please see the link
The MRR score on the test and validation sets. The higher, the better.
Package: >=0.8.0
| Rank |
Method |
Test NDCG@10 |
Validation NDCG@10 |
Contact |
References |
Date |
| 1 |
Persistent Forecast |
0.855 |
0.860 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 2 |
Moving Average |
0.823 |
0.841 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 3 |
TGNv2 |
0.735 |
0.807 |
Benedict Aaron Tjandra(University of Oxford) |
Paper, Code |
November 15th, 2024 |
| 4 |
DyGFormer |
0.388 ± 0.006 |
0.408 ± 0.006 |
Le Yu (Beihang University) |
Paper, Code |
August 22nd, 2023 |
| 5 |
TGN |
0.374 ± 0.001 |
0.395 ± 0.002 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 6 |
DyRep |
0.374 ± 0.001 |
0.394 ± 0.001 |
TGB team |
Paper, Code |
August 6th, 2023 |
The MRR score on the test and validation sets. The higher, the better.
Package: >=0.8.0
| Rank |
Method |
Test NDCG@10 |
Validation NDCG@10 |
Contact |
References |
Date |
| 1 |
Moving Average |
0.509 |
0.499 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 2 |
TGNv2 |
0.469 |
0.481 |
Benedict Aaron Tjandra(University of Oxford) |
Paper, Code |
November 15th, 2024 |
| 3 |
TGN |
0.367 ± 0.058 |
0.403 ± 0.010 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 4 |
DyGFormer |
0.365 ± 0.002 |
0.371 ± 0.001 |
Le Yu (Beihang University) |
Paper, Code |
August 22nd, 2023 |
| 5 |
Persistent Forecast |
0.357 |
0.350 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 6 |
DyRep |
0.351 ± 0.001 |
0.357 ± 0.001 |
TGB team |
Paper, Code |
August 6th, 2023 |
The MRR score on the test and validation sets. The higher, the better.
Package: >=0.8.0
| Rank |
Method |
Test NDCG@10 |
Validation NDCG@10 |
Contact |
References |
Date |
| 1 |
Moving Average |
0.559 |
0.574 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 2 |
TGNv2 |
0.507 |
0.544 |
Benedict Aaron Tjandra(University of Oxford) |
Paper, Code |
November 15th, 2024 |
| 3 |
Persistent Forecast |
0.369 |
0.380 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 4 |
DyGFormer |
0.316 ± 0.001 |
0.348 ± 0.001 |
Le Yu (Beihang University) |
Paper, Code |
August 22nd, 2023 |
| 5 |
TGN |
0.315 ± 0.020 |
0.379 ± 0.004 |
TGB team |
Paper, Code |
August 6th, 2023 |
| 6 |
DyRep |
0.312 ± 0.001 |
0.344 ± 0.001 |
TGB team |
Paper, Code |
August 6th, 2023 |
The MRR score on the test and validation sets. The higher, the better.
Package: >=0.9.0
| Rank |
Method |
Test NDCG@10 |
Validation NDCG@10 |
Contact |
References |
Date |
| 1 |
Moving Average |
0.508 |
0.491 |
TGB team |
Paper, Code |
September 27th, 2023 |
| 2 |
Persistent Forecast |
0.430 |
0.403 |
TGB team |
Paper, Code |
September 27th, 2023 |
| 3 |
TGNv2 |
0.294 |
0.321 |
Benedict Aaron Tjandra(University of Oxford) |
Paper, Code |
November 15th, 2024 |
| 4 |
TGN |
0.169 ± 0.006 |
0.189 ± 0.005 |
TGB team |
Paper, Code |
September 27th, 2023 |
| 5 |
DyRep |
0.141 ± 0.006 |
0.151 ± 0.006 |
TGB team |
Paper, Code |
September 27th, 2023 |