Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Calibration on Average (StrategyQA, HotpotQA, NQ, Bamboogle)

0.264ECE

NAACL

0.253880.322190.39050.45881Jan 16, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
0.2640.768
2026.01
0.2660.751
2026.01
0.3110.679
2026.01
0.3140.667
2026.01
0.3150.722
2026.01
0.3230.697
2026.01
0.3350.726
2026.01
0.3440.723
2026.01
0.3520.648
2026.01
0.3530.686
2026.01
0.360.722
2026.01
0.3660.73
2026.01
0.3730.615
2026.01
0.3770.591
2026.01
0.3880.571
2026.01
0.4070.657
2026.01
0.4070.648
2026.01
0.4110.605
2026.01
0.4360.65
2026.01
0.4370.654
2026.01
0.440.603
2026.01
0.4620.623
2026.01
0.4740.655
2026.01
0.5170.573