Bamboogle
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Bamboogle out-of-domain (test)
68.8Accuracy (R)
14
Bamboogle 500 samples (val)
34.6EM
14
Bamboogle
37.1OSR
13
Bamboogle out-of-domain
72Accuracy
12
Bamboogle
58.7CEM (%)
12
Bamboogle
45.6Accuracy
12
Bamboogle
64.8LJFT Score
12
Bamboogle
77.6Success Rate
12
Bamboogle Wiki18 index (test)
64.3Pass@4
10
Bamboogle
48.8Exact Match (EM)
10
Bamboogle
62Accuracy
10
Bamboogle multi-hop (test)
40.1Avg@16
10
Bamboogle auto-eval (test)
76.1Mean Accuracy
10
Bamboogle (test)
75.2Accuracy
9
Bamboogle (test)
59.2Pass@3
8
Bamboogle
0.521ECE
8
Bamboogle Wikipedia-18 retriever context (test)
40.32EM
7
Bamboogle
67.89LLM Judge Score
6
Bamboogle out-of-domain (test)
34.4Accuracy
6
Bamboogle 5Q
14.9Exact Match
6
Bamboogle 3Q
30.6Exact Match
6
Bamboogle
30.4Pass@1
6
Bamboogle
77.3Accuracy
6
Bamboogle
50.4subEM
5
Bamboogle
53.61F1 Score
5