BoolQ
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BoolQ
37AUC π-Soft-NS
27
BoolQ
-0.01Delta Accuracy
24
BoolQ (test)
67.4Accuracy
22
BoolQ In-Distribution
90.5Accuracy
21
BoolQ
0.23Loss
20
BoolQ
88.7Acc (Normalized)
20
BoolQ M
100Rk
20
BoolQ HELM
89.75Balanced Accuracy
18
BoolQ
86.1Calibrated Accuracy
18
BoolQ
77.68Accuracy
17
BoolQ
91.7Accuracy
16
BoolQ (test)
20.78Sufficiency
16
BoolQ
2.16Delta Accuracy
15
BoolQ
85.22Accuracy (Neutral)
15
BoolQ (dev)
84.8Accuracy
15
BoolQ
67Accuracy (Inter-Layer Filtering)
15
BoolQ-NP
73.41Accuracy
14
BoolQ
84.7BoolQ Score
12
BoolQ
100Accuracy
12
BoolQ (test)
79.2Accuracy
12
BoolQ
1.4F1 Score
10
BoolQ
71.38Character-level Accuracy
10
BoolQ
69.7Score
10
BoolQ (val)
90.9Accuracy
10
BoolQ SuperGLUE (val)
78.57Accuracy
9