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

Performance Prediction on MMLU 14k (test)

1.07MAE

DISCO

0.97481.61742.262.9026Oct 9, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
1.070.987
2025.10
1.140.975
2025.10
1.30.987
2025.10
1.310.972
2025.10
1.810.933
2025.10
1.820.912
2025.10
2.080.927
2025.10
2.080.904
2025.10
2.790.922
2025.10
3.250.922
2025.10
3.450.916