Deductive Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
99.2Pass@1
38
Apr 23, 2026
92.5Accuracy
28
Apr 23, 2026
99.67End-to-end Accuracy
21
Mar 19, 2026
59.05Accuracy
18
Apr 28, 2026
97.4Pass@1
18
Feb 26, 2026
100Accuracy
17
Mar 19, 2026
0.22Normalized Generalization Error Approximation
10
Jun 15, 2026
92.1Calibrated Accuracy
8
Feb 26, 2026