Bamboogle
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Bamboogle
73.8F1
23
Bamboogle
68.4Success Rate
22
Bamboogle
57.6EM
21
Bamboogle
49.6EM
20
Bamboogle Wiki-18
53.4Exact Match (EM)
20
Bamboogle
65.2Token-Level F1
20
Bamboogle standard (val)
40Exact Match (EM)
20
Bamboogle (out-of-domain)
73.8Accuracy
19
Bamboogle out-of-domain
46.4Exact Match
18
Bamboogle
69.6pass@1
18
Bamboogle
42.4EM
18
Bamboogle
62.4Cover Exact Match
18
Bamboogle
72SR
17
Bamboogle (dev test)
68.2F1 Score
17
Bamboogle
44.56EM
16
Bamboogle
50Score
16
Bamboogle OOD
44.6Exact Match Accuracy
15
Bamboogle
60.4F1 Score
15
Bamboogle out-of-domain (val test)
56.4Exact Match (EM)
15
Bamboogle
18.4Joint Answer-Evidence Score
14
Bamboogle
45.5Exact Match (EM)
14
Bamboogle
62.25F1 Score
14
Bamboogle
73.1String-F1 Score
14
Bamboogle 125 QA pairs
45.9EM
14
Bamboogle 2018 Wikipedia dump (dev)
45.2Accuracy
14