CommonsenseQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
CommonsenseQA
64.2Accuracy
19
CommonsenseQA
0.74AUROC
18
CommonsenseQA (dev)
76.2Accuracy
18
CommonsenseQA ConceptNet (20% test)
81.3Accuracy
16
CommonsenseQA
0.595PR-AUC
16
CommonSenseQA n=1700 (20%)
68.12Accuracy
13
CommonSenseQA n=425 (5%)
61.22Accuracy
13
CommonsenseQA
74Accuracy
12
CommonsenseQA
85.18Accuracy
10
CommonsenseQA (first 300 questions)
64.44Average Accuracy
10
CommonsenseQA
15.54LLMcritic Calls
10
CommonsenseQA
1Attack Success Rate (ASR)
9
CommonsenseQA (dev)
73.58Accuracy (5%)
9
CommonsenseQA 3k (dev)
71.52Accuracy (5%)
9
CommonSenseQA
85.5Pass@1 Accuracy
8
CommonsenseQA (val)
82.8Accuracy
8
CommonsenseQA (CSQA2) 2.0 (dev)
58.72Accuracy
8
CommonsenseQA (CQA) (test)
51.9AUC
7
CommonSenseQA FS n=30
52.5Accuracy
6
CommonsenseQA Multiple Client (test)
80.61Client 1 Accuracy
6
CommonsenseQA Single Client
83.78Accuracy
6
CommonsenseQA
39.12Score
6
CommonsenseQA (out-of-domain)
70.7Accuracy
5
CommonsenseQA (CQA)
85.09Accuracy
5
CommonsenseQA bn
26Normalized Accuracy
4