Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Input OOD Detection on Samsum
Loading...
1
AUROC
Deep SAD
0.987936
0.991068
0.9942
0.997332
Feb 5, 2026
AUROC
FPR@95
Updated 4d ago
Evaluation Results
Method
Method
Links
AUROC
FPR@95
Deep SAD
Backbone=PEGASUS_LARGE...
2026.02
1
0
Relative Mahalanobis
Backbone=PEGASUS_LARGE...
2026.02
0.9999
0.01
AP-OOD
Backbone=PEGASUS_LARGE...
2026.02
0.9999
0
Binary logits
Backbone=PEGASUS_LARGE...
2026.02
0.9998
0.03
Mahalanobis
Backbone=PEGASUS-LARGE...
2026.02
0.9977
0.17
AP-OOD
Backbone=PEGASUS-LARGE...
2026.02
0.9976
0
Deep SVDD
Backbone=PEGASUS-LARGE...
2026.02
0.9957
0.72
KNN
Backbone=PEGASUS-LARGE...
2026.02
0.9884
3.09
Feedback
Search any
task
Search any
task