PDF

Reported on 4 benchmarks across 3 tasks · 2 papers

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Graphs2 results

Graph RegressiononZINC-500k
MAE· 2023-05-10
0.066
best: 0.051 (ESA + rings + NodeRWSE + EdgeRWSE)
Towards Better Graph Representation Learning with Parameterized Decomposition & Filtering arXiv:2305.06102
Graph Property Predictiononogbg-molpcba
Number of params· 2023-05-10
3842048
best: 119529665 (HIG(pre-trained on PCQM4M))
Towards Better Graph Representation Learning with Parameterized Decomposition & Filtering arXiv:2305.06102

Computer Vision2 results

Person Re-IdentificationonMarket-1501
Rank-1· 2017-09-25
84.14
best: 98 (st-ReID(RE, RK))
Pose-driven Deep Convolutional Model for Person Re-identification arXiv:1709.08325
Person Re-IdentificationonMarket-1501
mAP· 2017-09-25
63.41
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
Pose-driven Deep Convolutional Model for Person Re-identification arXiv:1709.08325