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/HSIC-Attribution

HSIC-Attribution

Reported on 15 benchmarks across 3 tasks · 1 paper · 8 SOTA

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

Reasoning8 results

Error UnderstandingonCUB-200-2011 (ResNet-101)
Average highest confidence· 2022-06-13
0.2493
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011 (ResNet-101)
Insertion AUC score· 2022-06-13
0.1446
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011
Average highest confidence (MobileNetV2)· 2022-06-13
0.2914
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011
Average highest confidence (ResNet-101)· 2022-06-13
0.2493
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011
Insertion AUC score (EfficientNetV2-M)· 2022-06-13
0.1611
best: 0.1748 (SMDL-Attribution (ICLR version))
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011
Insertion AUC score (MobileNetV2)· 2022-06-13
0.1635
best: 0.1922 (SMDL-Attribution (ICLR version))
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011
Insertion AUC score (ResNet-101)· 2022-06-13
0.1446
best: 0.1772 (SMDL-Attribution (ICLR version))
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Error UnderstandingonCUB-200-2011
Average highest confidence (EfficientNetV2-M)· 2022-06-13
0.2679
best: 0.2403 (Score-CAM)
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219

Computer Vision6 results

Image AttributiononVGGFace2
Insertion AUC score (ArcFace ResNet-101)· 2022-06-13
0.6694
best: 0.6705 (SMDL-Attribution (ICLR version))
SOTA
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Image AttributiononVGGFace2
Deletion AUC score (ArcFace ResNet-101)· 2022-06-13
0.1317
best: 0.3103 (Grad-CAM)
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Image AttributiononCUB-200-2011
Deletion AUC score (ResNet-101)· 2022-06-13
0.0647
best: 0.107 (LIME)
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Image AttributiononCUB-200-2011
Insertion AUC score (ResNet-101)· 2022-06-13
0.6843
best: 0.7262 (SMDL-Attribution (ICLR version))
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Image AttributiononCelebA
Deletion AUC score (ArcFace ResNet-101)· 2022-06-13
0.1151
best: 0.2865 (Grad-CAM)
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219
Image AttributiononCelebA
Insertion AUC score (ArcFace ResNet-101)· 2022-06-13
0.5692
best: 0.5752 (SMDL-Attribution (ICLR version))
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219

Miscellaneous1 result

Interpretability Techniques for Deep LearningonCelebA
Insertion AUC score· 2022-06-13
0.5692
best: 0.5703 (RISE)
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure arXiv:2206.06219