Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Domain Shift Extractive Question Answering on AdversarialQA (test)
Loading...
0.075
ECE
SBA
0.07232
0.09041
0.1085
0.12659
Feb 19, 2026
ECE
Selection AUROC
Updated 4d ago
Evaluation Results
Method
Method
Links
ECE
Selection AUROC
SBA
Backbone=LLaMA-2-7B
2026.02
0.075
0.782
Deep-Ens
Backbone=LLaMA-2-7B
2026.02
0.088
0.756
Temp. Scal.
Backbone=LLaMA-2-7B
2026.02
0.097
-
Gauss+Proj
Backbone=LLaMA-2-7B
2026.02
0.102
0.729
Laplace-LoRA
Backbone=LLaMA-2-7B
2026.02
0.108
0.718
LoRA
Backbone=LLaMA-2-7B
2026.02
0.142
0.658
Feedback
Search any
task
Search any
task