Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Knowledge Gap Detection on MMLU

78.2Accuracy

Align-P

45.9654.3362.771.07Apr 3, 2026
Updated 13d ago

Evaluation Results

MethodLinks
2026.04
78.20.534
2026.04
76.80.838
2026.04
75.90.506
2026.04
75.50.476
2026.04
750.809
2026.04
73.50.688
2026.04
73.40.758
2026.04
72.70.778
2026.04
71.30.678
2026.04
70.30.751
2026.04
70.20.765
2026.04
70.10.728
2026.04
51.9-
2026.04
50.50.567
2026.04
50.40.651
2026.04
50.20.536
2026.04
49-
2026.04
47.2-