CommonsenseQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
CommonsenseQA
1Power
207
CommonsenseQA
89.3Accuracy
172
CommonSenseQA
91.2Accuracy
136
CommonsenseQA (CSQA)
91.2Accuracy
124
CommonsenseQA
86.6Accuracy (pass@1)
108
CommonSenseQA
88.9Accuracy
92
CommonsenseQA (CSQA)
85.7Accuracy
62
CommonsenseQA
0.756Mean AUROC
62
CommonsenseQA (test)
90Accuracy
62
CommonsenseQA (test)
83.3Accuracy
60
CommonsenseQA IH (test)
88.9Accuracy
57
CommonSenseQA
0.105BS
54
CommonsenseQA IH (dev)
82.7Accuracy
53
CommonsenseQA (val)
82.06Accuracy
52
CommonsenseQA
74.48AUC
51
CommonsenseQA (CSQA) v1.0 (test)
64.11Accuracy
46
CommonsenseQA Non-Math
87.31Accuracy
32
CommonsenseQA
86.81Accuracy
25
CommonsenseQA (CSQA) (val)
75.7Accuracy
23
CommonsenseQA v1.0 (dev)
79.3Accuracy
22
CommonsenseQA (CSQA) (test)
40.85VWR
21
CommonsenseQA (CSQA)
66.4Accuracy
21
COMMONSENSEQA 1,000 examples
0.935Mean Hamming Similarity
20
CommonSenseQA CoQA
66.91Score
20
CommonsenseQA blind v1.0 (test)
75.3Accuracy
20