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

Multiple-choice Question Answering on MMLU (zero-shot, test)

76Accuracy (MMLU zero-shot)

SFT

18.2833.26548.2563.235Mar 10, 2026Mar 22, 2026Apr 3, 2026Apr 15, 2026Apr 27, 2026May 9, 2026May 22, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.03
76-
2026.03
75-
2026.03
75-
2026.03
73-
2026.03
72.5-
68.60.74
2026.03
68.17-
2026.03
68-
2026.05
67.86-
2026.03
67.67-
2026.03
66.17-
2026.03
63.5-
2026.03
63.17-
2026.03
60.5-
2026.03
59.33-
2026.03
57.83-
2026.03
56-
2026.03
53.83-
2026.03
53.83-
2026.03
53.5-
2026.03
47.33-
2026.03
44.5-
2026.03
34.17-
2026.03
32.83-
2026.03
27.5-
2026.03
24.5-
2026.03
20.5-