Logical reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.335Answer Accuracy
3
Feb 26, 2026
31Accuracy
3
Feb 27, 2026
23.1Accuracy
2
Jul 7, 2026
88.8Accuracy (OpenBookQA Test)
2
Jun 15, 2026
38AR-LSAT pass@1
2
Jun 2, 2026
69.24Pass@1
2
Jun 2, 2026
73.23Pass@1
2
Jun 2, 2026
92.26Accuracy
2
May 11, 2026
92.81Accuracy
2
May 11, 2026
100Accuracy under Attack
2
Apr 28, 2026
45.6Comprehensive Score
2
Apr 10, 2026
34.72in
2
Feb 26, 2026
36.42-Intersection
2
Feb 26, 2026
30.11Normalized Accuracy
2
Feb 26, 2026
48Accuracy
2
Feb 26, 2026
85.5Accuracy
2
Feb 26, 2026
85.9Accuracy
2
Feb 26, 2026
97.8OA (%)
1
Feb 26, 2026
99.8OA
1
Feb 26, 2026
2.7Accuracy
1
Feb 26, 2026