Explanation Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
13.19Sufficiency
16
Feb 26, 2026
9.72Sufficiency
16
Feb 26, 2026
7.08Sufficiency
16
Feb 26, 2026
20.78Sufficiency
16
Feb 26, 2026
43.76Sufficiency
16
Feb 26, 2026
17.69Sufficiency
14
Feb 26, 2026
0.39Deletion Score (Zero)
12
Feb 26, 2026
8.01Deletion Fidelity (Zero-shot)
12
Feb 26, 2026
26.21Sag (K=10)
6
Feb 26, 2026
23.92Sag (K=10)
6
Feb 26, 2026
20.4IoU (Masks)
3
Mar 25, 2026
74.2CH (%)
2
Feb 26, 2026