Commonsense Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
74.94CSQA Accuracy
3
May 15, 2026
81.61Performance
3
May 12, 2026
75.3Performance Score
3
May 12, 2026
43.9Accuracy
3
Feb 26, 2026
66.1Average QA Score
3
Feb 26, 2026
0.676Score
2
May 15, 2026
75ACC
2
May 14, 2026
100Accuracy under Attack
2
Apr 28, 2026
11.75ECE
2
Feb 26, 2026
57.2Performance Score (Finetune Baseline vs Predict Baseline)
2
Feb 26, 2026
84.9Accuracy
2
Feb 26, 2026