Zero-shot Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
78.24PIQA Accuracy
56
May 21, 2026
69.39WinoGrande Accuracy
35
Jun 15, 2026
57.7Accuracy
25
Feb 26, 2026
83.76BoolQ Accuracy (Zero-shot)
21
Mar 9, 2026
43.43ARC-C Accuracy
17
Mar 4, 2026
58.4ARC-c Accuracy (Zero-shot)
15
Feb 26, 2026
54.6Accuracy
12
Apr 24, 2026
46.8Accuracy on OBQA (Zero-shot)
8
Jun 9, 2026
82.7PIQA Accuracy (Zero-shot)
7
Apr 16, 2026
28.4Accuracy
5
May 6, 2026
33.53Accuracy
5
May 6, 2026
60.15ARC-C Accuracy
3
Jun 11, 2026
1.03Average Loss
2
Feb 26, 2026