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

Failure Diagnosis on MMLU

34Macro Similarity Type

PROBELLM

13.218.62429.4Feb 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
346631366
2026.02
326827295
2026.02
2773817419
2026.02
2674767516
2026.02
2377868116
2026.02
1981817820
2026.02
1684716310
2026.02
1486595716