Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/GAT

GAT

Reported on 34 benchmarks across 19 tasks · 5 papers · 20 SOTA

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

Graphs16 results

Graph RegressiononLipophilicity
RMSE· 2017-10-30
0.95
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonPPI
F1· 2017-10-30
97.3
best: 99.71 (g2-MLP)
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonPubmed
F1-Score· 2017-10-30
79
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonIMDB (Heterogeneous Node Classification)
Macro-F1· 2017-10-30
58.94
best: 66.63 (SeHGNN)
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonIMDB (Heterogeneous Node Classification)
Micro-F1· 2017-10-30
64.86
best: 69.77 (RpHGNN)
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonFreebase (Heterogeneous Node Classification)
Accuracy· 2017-10-30
65.26
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonFreebase (Heterogeneous Node Classification)
Macro-F1· 2017-10-30
40.74
best: 50.71 (SeHGNN)
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonDBLP (Heterogeneous Node Classification)
Macro-F1· 2017-10-30
93.83
best: 94.86 (SeHGNN)
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonDBLP (Heterogeneous Node Classification)
Micro-F1· 2017-10-30
93.39
best: 95.55 (RpHGNN)
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonACM (Heterogeneous Node Classification)
Macro-F1· 2017-10-30
92.26
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonACM (Heterogeneous Node Classification)
Micro-F1· 2017-10-30
92.19
best: 94.06 (SlotGAT)
SOTA
Graph Attention Networks arXiv:1710.10903
Node Property Predictiononogbn-proteins
Number of params· 2024-06-13
2943472
best: 664233700 (LD+GAT)
Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification arXiv:2406.08993
Graph RegressiononZINC 100k
MAE· 2017-10-30
0.463
best: 0.094 (CIN-small)
Graph Attention Networks arXiv:1710.10903
Graph ClassificationonCIFAR10 100k
Accuracy (%)· 2017-10-30
65.48
best: 76.468 (GRIT)
Graph Attention Networks arXiv:1710.10903
Node ClassificationonPATTERN 100k
Accuracy (%)· 2017-10-30
75.824
best: 86.816 (EGT)
Graph Attention Networks arXiv:1710.10903
Graph Property Predictiononogbg-code2
Number of params· 2017-10-30
11030210
best: 63684290 (GMAN+bag of tricks)
Graph Attention Networks arXiv:1710.10903

Computer Vision4 results

Image Classificationon75 Superpixel MNIST
Classification Error· 2020-02-13
3.81
best: 0.95 (Dynamic Reduction Network (256 HD))
SOTA
Superpixel Image Classification with Graph Attention Networks arXiv:2002.05544
VideoonJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903
Temporal Action LocalizationonJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903
Action LocalizationonJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903

Miscellaneous4 results

Fraud DetectiononElliptic Dataset
AUC· 2024-05-29
0.8102
best: 0.8329 (GCN)
Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental Evaluation arXiv:2405.19383
Fraud DetectiononElliptic Dataset
AUPRC· 2024-05-29
0.5886
best: 0.6312 (GraphSAGE)
Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental Evaluation arXiv:2405.19383
Twitter Bot DetectiononMGTAB
Acc· 2023-01-03
87
best: 92.1 (RGT)
MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark arXiv:2301.01123
Twitter Bot DetectiononMGTAB
F1· 2023-01-03
82.3
best: 90.4 (RGT)
MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark arXiv:2301.01123

Robots3 results

Activity RecognitiononJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903
Active Speaker DetectiononElliptic Dataset
AUC· 2024-05-29
0.8102
best: 0.8329 (GCN)
Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental Evaluation arXiv:2405.19383
Active Speaker DetectiononElliptic Dataset
AUPRC· 2024-05-29
0.5886
best: 0.6312 (GraphSAGE)
Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental Evaluation arXiv:2405.19383

Natural Language Processing3 results

3D Action RecognitiononJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903
Stance DetectiononMGTAB
Acc· 2023-01-03
82.2
best: 87.8 (RGT)
MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark arXiv:2301.01123
Stance DetectiononMGTAB
F1· 2023-01-03
81
best: 86.9 (RGT)
MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark arXiv:2301.01123

Methodology2 results

Zero-Shot LearningonJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903
ClassificationonCIFAR10 100k
Accuracy (%)· 2017-10-30
65.48
best: 76.468 (GRIT)
Graph Attention Networks arXiv:1710.10903

Time Series2 results

Action DetectiononJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903
Action RecognitiononJ-HMBD Early Action
10%· 2017-10-30
58.1
best: 60.6 (DR^2N)
SOTA
Graph Attention Networks arXiv:1710.10903