Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Correctness Prediction on MMLU

65.83Accuracy

IRT-NET

51.5355.242558.95562.6675Jan 28, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
65.83
2026.01
65.62
2026.01
65.55
2026.01
65.39
2026.01
65.2
2026.01
65.19
2026.01
64.8
2026.01
64.64
2026.01
64.54
2026.01
64.16
2026.01
64.01
2026.01
63.82
2026.01
63.57
2026.01
63.25
2026.01
62.99
2026.01
62.77
2026.01
61.33
2026.01
52.08