Paired-prompt evaluation
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
58.81Simple Accuracy
2
Feb 26, 2026
64.56Simple Accuracy
2
Feb 26, 2026
67.81Simple Accuracy
2
Feb 26, 2026
75.7Simple Accuracy
2
Feb 26, 2026
52.89Simple Accuracy
2
Feb 26, 2026
90.67Simple Accuracy
2
Feb 26, 2026