Commonsense Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
89.45Accuracy
35
Jul 3, 2026
74.5Accuracy
35
Apr 23, 2026
87.3Accuracy
35
Mar 18, 2026
55.58Average Accuracy
34
Jun 2, 2026
82.53Accuracy
34
Feb 26, 2026
84.9Avg Commonsense Accuracy
34
Feb 26, 2026
97.6Accuracy
34
Feb 26, 2026
82.1Accuracy
32
Mar 4, 2026
89.9PIQA Accuracy
32
Feb 27, 2026
87.31Accuracy
32
Feb 26, 2026
88.2Overall Average Accuracy
31
Jul 2, 2026
82.94Accuracy (Baseline)
31
Feb 26, 2026
74.1Accuracy
31
Apr 21, 2026
82.5Task Success Rate (TSR)
30
May 28, 2026
74.6RCC
29
Jun 12, 2026
70.87Accuracy
28
Apr 8, 2026
60Overall Average Accuracy
28
Feb 26, 2026
62.5Average Accuracy
28
Jun 18, 2026
75.1BoolQ Accuracy
27
May 12, 2026
77.4MV Score
27
Apr 3, 2026
83.3Score
27
Apr 28, 2026
89.3Accuracy
27
Apr 14, 2026
75.24BoolQ Accuracy
26
Jun 16, 2026
81.35BoolQ Accuracy
26
May 19, 2026
53.27Normalized PLL Score
26
Mar 4, 2026