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/CSI

CSI

Reported on 7 benchmarks across 1 task · 1 paper · 6 SOTA

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

Methodology7 results

  • Anomaly DetectiononOne-class ImageNet-30
    AUROC· 2020-07-16
    91.6
    best: 99.9 (BCE-Clip (OE))
    SOTA
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176
  • Anomaly DetectiononAnomaly Detection on Anomaly Detection on Unlabeled ImageNet-30 vs Flowers-102
    ROC-AUC· 2020-07-16
    94.7
    best: 98.3 (PsudoLabels CLIP ViT)
    SOTA
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176
  • Anomaly DetectiononAnomaly Detection on Unlabeled CIFAR-10 vs LSUN (Fix)
    ROC-AUC· 2020-07-16
    90.3
    best: 99.1 (PsudoLabels ViT)
    SOTA
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176
  • Anomaly DetectiononUnlabeled CIFAR-10 vs CIFAR-100
    AUROC· 2020-07-16
    89.3
    best: 96.7 (PsudoLabels ViT)
    SOTA
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176
  • Anomaly DetectiononOne-class CIFAR-100
    AUROC· 2020-07-16
    89.6
    best: 98.4 (GeneralAD)
    SOTA
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176
  • Anomaly DetectiononOne-class CIFAR-10
    AUROC· 2020-07-16
    94.3
    best: 99.6 (CLIP (OE))
    SOTA
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176
  • Anomaly DetectiononAnomaly Detection on Unlabeled ImageNet-30 vs CUB-200
    ROC-AUC· 2020-07-16
    71.5
    best: 99.4 (PsudoLabels CLIP ViT)
    CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted InstancesarXiv:2007.08176