Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Image Classification on CIFAR-100 (Robustness and Calibration Metrics)
Loading...
5.58
Average Accuracy
Reveal-path Objective
0.744
1.9995
3.255
4.5105
May 8, 2026
Average Accuracy
Unseen Data Performance
OOD Performance
Path Sensitivity (Logit)
Path Sensitivity (Hidden)
Expected Calibration Error (ECE)
Updated 22d ago
Evaluation Results
Method
Method
Links
Average Accuracy
Unseen Data Performance
OOD Performance
Path Sensitivity (Logit)
Path Sensitivity (Hidden)
Expected Calibration Error (ECE)
Reveal-path Objective
Objective=reveal
2026.05
5.58
4.9
6
7.2521
3.3842
2.39
Task Objective
Objective=task
2026.05
4.05
2.95
5.7
15.1867
7.2761
4.73
Full Field Objective
Objective=full
2026.05
0.93
0.9
1
3.0355
2.1391
2.29
Feedback
Search any
task
Search any
task