Selective Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0Fail Rate
70
Jul 7, 2026
2,849Score (Coverage 14%)
40
May 20, 2026
92.5Area under Coverage-F1
12
Feb 26, 2026
92.5AUC-F1 (Coverage)
12
Feb 26, 2026
94.5Area under Coverage-F1
12
Feb 26, 2026
13Accuracy
11
May 29, 2026
21Accuracy
11
May 29, 2026