Robustness Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.57Mean Normalized Performance
70
Jun 4, 2026
1.603Normalized Performance
70
Jun 4, 2026
66.5R@1
31
Feb 26, 2026
58R@1
31
Feb 26, 2026
48.3R@1
31
Feb 26, 2026
0.02FPR (%)
20
Feb 26, 2026
0FPR (%)
20
Feb 26, 2026
0FPR (%)
20
Feb 26, 2026
0FPR
20
Feb 26, 2026
74.17Overall Accuracy (OA)
12
May 13, 2026
73.53MCQ Accuracy (low)
9
Jun 9, 2026
100Identity Bit Accuracy
9
Feb 26, 2026
100Identity Bit Acc
9
Feb 26, 2026
62.56Change Accuracy
8
Mar 6, 2026
83.07Accuracy
8
Feb 26, 2026
82.5Accuracy
8
Feb 26, 2026
0.048Jensen-Shannon Divergence
8
Feb 26, 2026
43.15mCE
8
Feb 26, 2026
70.9Accuracy
6
Apr 24, 2026
72.9Accuracy
6
Apr 24, 2026
33.5GACC
6
Apr 24, 2026
231.8Average Reward
6
Feb 26, 2026
830.9Average Reward
6
Feb 26, 2026
810.1Average Reward
6
Feb 26, 2026
1,000Average Reward
6
Feb 26, 2026