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/LWS

LWS

Reported on 40 benchmarks across 5 tasks · 1 paper · 35 SOTA

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

Methodology24 results

  • Generalized Few-Shot ClassificationonCIFAR-10-LT (ρ=10)
    Error Rate· 2019-10-21
    8.9
    best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonCUB-LT
    Long-Tailed Accuracy· 2019-10-21
    65.7
    best: 67.7 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonCUB-LT
    Per-Class Accuracy· 2019-10-21
    53.1
    best: 60.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonAWA-LT
    Per-Class Accuracy· 2019-10-21
    73.4
    best: 76.2 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonSUN-LT
    Long-Tailed Accuracy· 2019-10-21
    40.2
    best: 40.4 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonSUN-LT
    Per-Class Accuracy· 2019-10-21
    33.9
    best: 36.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonImageNet-LT-d
    Per-Class Accuracy· 2019-10-21
    49.9
    best: 53.5 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonCIFAR-10-LT (ρ=10)
    Error Rate· 2019-10-21
    8.9
    best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonCUB-LT
    Long-Tailed Accuracy· 2019-10-21
    65.7
    best: 67.7 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonCUB-LT
    Per-Class Accuracy· 2019-10-21
    53.1
    best: 60.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonAWA-LT
    Per-Class Accuracy· 2019-10-21
    73.4
    best: 76.2 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonSUN-LT
    Long-Tailed Accuracy· 2019-10-21
    40.2
    best: 40.4 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonSUN-LT
    Per-Class Accuracy· 2019-10-21
    33.9
    best: 36.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonImageNet-LT-d
    Per-Class Accuracy· 2019-10-21
    49.9
    best: 53.5 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonCIFAR-10-LT (ρ=10)
    Error Rate· 2019-10-21
    8.9
    best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonCUB-LT
    Long-Tailed Accuracy· 2019-10-21
    65.7
    best: 67.7 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonCUB-LT
    Per-Class Accuracy· 2019-10-21
    53.1
    best: 60.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonAWA-LT
    Per-Class Accuracy· 2019-10-21
    73.4
    best: 76.2 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonSUN-LT
    Long-Tailed Accuracy· 2019-10-21
    40.2
    best: 40.4 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonSUN-LT
    Per-Class Accuracy· 2019-10-21
    33.9
    best: 36.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonImageNet-LT-d
    Per-Class Accuracy· 2019-10-21
    49.9
    best: 53.5 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot ClassificationonAWA-LT
    Long-Tailed Accuracy· 2019-10-21
    93.5
    best: 94.1 (DRAGON)
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Long-tail LearningonAWA-LT
    Long-Tailed Accuracy· 2019-10-21
    93.5
    best: 94.1 (DRAGON)
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Generalized Few-Shot LearningonAWA-LT
    Long-Tailed Accuracy· 2019-10-21
    93.5
    best: 94.1 (DRAGON)
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217

Computer Vision16 results

  • Image ClassificationonCIFAR-10-LT (ρ=10)
    Error Rate· 2019-10-21
    8.9
    best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonCUB-LT
    Long-Tailed Accuracy· 2019-10-21
    65.7
    best: 67.7 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonCUB-LT
    Per-Class Accuracy· 2019-10-21
    53.1
    best: 60.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonAWA-LT
    Per-Class Accuracy· 2019-10-21
    73.4
    best: 76.2 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonSUN-LT
    Long-Tailed Accuracy· 2019-10-21
    40.2
    best: 40.4 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonSUN-LT
    Per-Class Accuracy· 2019-10-21
    33.9
    best: 36.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonImageNet-LT-d
    Per-Class Accuracy· 2019-10-21
    49.9
    best: 53.5 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonCIFAR-10-LT (ρ=10)
    Error Rate· 2019-10-21
    8.9
    best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonCUB-LT
    Long-Tailed Accuracy· 2019-10-21
    65.7
    best: 67.7 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonCUB-LT
    Per-Class Accuracy· 2019-10-21
    53.1
    best: 60.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonAWA-LT
    Per-Class Accuracy· 2019-10-21
    73.4
    best: 76.2 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonSUN-LT
    Long-Tailed Accuracy· 2019-10-21
    40.2
    best: 40.4 (DRAGON)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonSUN-LT
    Per-Class Accuracy· 2019-10-21
    33.9
    best: 36.1 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonImageNet-LT-d
    Per-Class Accuracy· 2019-10-21
    49.9
    best: 53.5 (DRAGON + Bal'Loss)
    SOTA
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Image ClassificationonAWA-LT
    Long-Tailed Accuracy· 2019-10-21
    93.5
    best: 94.1 (DRAGON)
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217
  • Few-Shot Image ClassificationonAWA-LT
    Long-Tailed Accuracy· 2019-10-21
    93.5
    best: 94.1 (DRAGON)
    Decoupling Representation and Classifier for Long-Tailed RecognitionarXiv:1910.09217