Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Post-hoc Explainability on Nine datasets average
Loading...
72.2
Macro-F1
GCE discrete
63.672
65.886
68.1
70.314
May 17, 2026
Macro-F1
Removal Drop
Prediction Gap
Completeness Degradation
Evidence Sufficiency
Updated 15d ago
Evaluation Results
Method
Method
Links
Macro-F1
Removal Drop
Prediction Gap
Completeness Degradation
Evidence Sufficiency
GCE discrete
Runtime=1.08×
2026.05
72.2
17.6
0.5
27.7
53.3
Occlusion top-k
Runtime=8.70×
2026.05
66.8
5.2
2.2
13.9
40.5
Integrated gradients
Runtime=3.80×
2026.05
66.2
4.8
2.3
12.8
39.6
Gradient saliency
Runtime=1.42×
2026.05
65.5
4.2
2.5
11.8
38.6
CLAM attention
Runtime=1.10×
2026.05
64.5
3.5
2.7
11.2
37.4
Attention top-k
Runtime=1.00×
2026.05
64
3.3
2.9
10.7
36.7
Feedback
Search any
task
Search any
task