BBH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BBH
9.26Brier Score (BBH)
11
BBH
66.5Logical Deduction Accuracy
11
BBH-Navigate (test)
98Accuracy
11
BBH
80.1BBH Score
10
bbh-zh
87.52Overall Score
10
BBH HHH
95Harmlessness Score
10
BBH-TSO
89.6BCA
9
BBH-LD
87.1BCA
9
BBH
77.45EM Score
9
BBH
77.4Best-of-K Accuracy
8
BBH
71.5T Score (BBH)
8
BBH 3-shot chain-of-thought
61.35EM
8
BBH
66.1Performance Score (BBH General Reasoning)
8
BBH
40.65BBH Comprehensive Reasoning Score
8
BBH
71.37MRS
7
BBH multiple-choice (first 1,000 samples)
86.2Exact Match Accuracy
7
BBH (test)
90.48AUROC
6
BBH
35.81BBH Complex Factual Reasoning Score
6
BBH subset
47.2Accuracy
6
BBH
52.5Exact Match Accuracy (BBH Logical Reasoning)
6
BBH
89.3pass@8
6
BBH
63.32Acc@8
6
BBH
70.6Normalized Log Accuracy
6
BBH Navigation (test)
83.1Accuracy
6
BBH
67.4BBH Solution Rate
6