TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

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

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

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

Models/GIN

GIN

Reported on 22 benchmarks across 8 tasks · 3 papers · 4 SOTA

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

Graphs11 results

  • Graph RegressiononPCQM4M-LSC
    Test MAE· 2021-03-17
    16.78
    best: 13.28 (Graphormer )
    SOTA
    OGB-LSC: A Large-Scale Challenge for Machine Learning on GraphsarXiv:2103.09430
  • Graph RegressiononPCQM4Mv2-LSC
    Test MAE· 2018-10-01
    0.1218
    best: 0.0683 (EGT + Triangular Attention)
    SOTA
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph RegressiononPCQM4Mv2-LSC
    Validation MAE· 2018-10-01
    0.1195
    best: 0.0235 (ESA (Edge set attention, no positional encodings))
    SOTA
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Node ClassificationonPATTERN 100k
    Accuracy (%)· 2018-10-01
    85.59
    best: 86.816 (EGT)
    SOTA
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph RegressiononZINC-500k
    MAE· 2018-10-01
    0.526
    best: 0.051 (ESA + rings + NodeRWSE + EdgeRWSE)
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph ClassificationonCIFAR10 100k
    Accuracy (%)· 2018-10-01
    53.28
    best: 76.468 (GRIT)
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph Property Predictiononogbg-molhiv
    Number of params· 2018-10-01
    1885206
    best: 47183040 (Graphormer)
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph Property Predictiononogbg-code2
    Number of params· 2018-10-01
    12390715
    best: 63684290 (GMAN+bag of tricks)
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph Property Predictiononogbg-ppa
    Number of params· 2018-10-01
    1836942
    best: 16346166 (PAS+F2GNN)
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph Property Predictiononogbg-molpcba
    Number of params· 2018-10-01
    1923433
    best: 119529665 (HIG(pre-trained on PCQM4M))
    How Powerful are Graph Neural Networks?arXiv:1810.00826
  • Graph Property Predictiononogbg-molhiv
    Number of params
    32385
    best: 47183040 (Graphormer)

Natural Language Processing7 results

  • Visual Question Answering (VQA)onGQA Test2019
    Accuracy
    56.16
    best: 89.3 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Binary
    73.56
    best: 91.2 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Consistency
    84.99
    best: 98.4 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Distribution
    5.87
    best: 93.08 (GlobalPrior)
  • Visual Question Answering (VQA)onGQA Test2019
    Open
    40.8
    best: 87.4 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Plausibility
    84.83
    best: 97.2 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Validity
    96.4
    best: 98.9 (human)

Miscellaneous2 results

  • Fraud DetectiononElliptic Dataset
    AUC· 2024-05-29
    0.8089
    best: 0.8329 (GCN)
    Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental EvaluationarXiv:2405.19383
  • Fraud DetectiononElliptic Dataset
    AUPRC· 2024-05-29
    0.5517
    best: 0.6312 (GraphSAGE)
    Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental EvaluationarXiv:2405.19383

Robots2 results

  • Active Speaker DetectiononElliptic Dataset
    AUC· 2024-05-29
    0.8089
    best: 0.8329 (GCN)
    Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental EvaluationarXiv:2405.19383
  • Active Speaker DetectiononElliptic Dataset
    AUPRC· 2024-05-29
    0.5517
    best: 0.6312 (GraphSAGE)
    Network Analytics for Anti-Money Laundering -- A Systematic Literature Review and Experimental EvaluationarXiv:2405.19383

Methodology1 result

  • ClassificationonCIFAR10 100k
    Accuracy (%)· 2018-10-01
    53.28
    best: 76.468 (GRIT)
    How Powerful are Graph Neural Networks?arXiv:1810.00826