BoolQ
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BoolQ 200-sample
83.5Accuracy
8
BoolQ Corrupt Setting
80.8Accuracy
8
BoolQ Clean Setting
89.4Accuracy
8
BoolQ
89.42Accuracy (BoolQ)
8
BoolQ
80Accuracy
8
BoolQ SuperGLUE (test dev)
90Accuracy
8
BoolQ SuperGLUE (test)
63.83Accuracy (Test)
8
BoolQ contrastive (test)
60.6Accuracy (Avg)
8
BoolQ (val)
87.5Accuracy
8
BoolQ
3.54Wall-clock Time (hours)
7
BoolQ
77.2Factual Accuracy (FA)
7
BoolQ
0.054MAE
7
BoolQ
61.93Accuracy
7
BoolQ
74.7Score (BoolQ)
6
BoolQ
71.07Accuracy
6
BoolQ
85.9Accuracy
6
BoolQ
74.1Accuracy
6
BoolQ fr
80.9Accuracy
6
NO-BoolQ NLEBench (test)
0.632Accuracy
6
BoolQ (abundant)
87.1Accuracy
6
BoolQ (medium)
85.7Accuracy
6
BoolQ (scarce)
79.1Accuracy
6
BoolQ
71.4Acc (0-shot)
6
BoolQ (dev)
86.1Accuracy
6
BoolQ CS
80.4Score
6