CB
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
CB
98.2Accuracy
129
CB
91.1Accuracy
76
CB SuperGLUE (test)
91.43Accuracy
33
CB
91Average Accuracy
29
CB
0.03Loss
20
CB val (test)
94.6Accuracy
19
CB
84.99Accuracy
16
CB (test)
89.3Accuracy
13
CB Far OOD
94.6Accuracy (CB Far OOD)
10
CB (test)
64.6Macro-F1
10
CB
5.43Total Communication Time (10^3 s)
9
CB EN
70.89MUSIQ Score
8
CB Human Interaction
48.3Success Rate
8
CB User Simulation
57.5Success Rate (SR)
8
CB SuperGLUE (test dev)
84Accuracy
8
CB
87.5Accuracy
8
CB (evaluation set)
59.08Precision
8
CB
61.51F1
7
CB (test)
6,894Number of Queries
6
CB UCI Repository (test)
74.8Accuracy
6
CB
0.537Macro F1 Score
6
CB
82.1Acc (0-shot)
6
CB (dev)
0.84Accuracy
6
CB 32 samples
86.5F1 Score
6
CB
87.5Final Accuracy
4