Reading Comprehension
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
86.1F1 Score
17
Apr 2, 2026
88.2F1 Score
17
Apr 21, 2026
74.33Accuracy
16
Jun 17, 2026
83.49Performance Score
16
Jun 11, 2026
28.32EM
16
Apr 21, 2026
95.4Accuracy (RACE MS)
16
Feb 27, 2026
77.38Accuracy
16
Feb 26, 2026
88.1Accuracy
16
Feb 26, 2026
91Score
15
May 8, 2026
0.128Mean Per-Step Regret
15
Feb 26, 2026
84.8Accuracy
15
Apr 1, 2026
0.922F1 Score
15
Feb 26, 2026
23.29EM
14
May 12, 2026
35.04EM
14
May 12, 2026
89.56Accuracy
14
May 12, 2026
91Accuracy
14
Apr 23, 2026
83.31Accuracy
14
Feb 26, 2026
43.62EM
14
Feb 26, 2026
0.488EM
14
Feb 26, 2026
45.5EM
14
Feb 26, 2026
67.62EM
14
Feb 26, 2026
83.36Vanilla Accuracy
14
Feb 26, 2026
58.64F1 Score
14
Feb 26, 2026
81.6EM
14
Feb 26, 2026
87.56Accuracy
13
Apr 14, 2026