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

Legal Reasoning on LegalBench (MAE)

0.033MAE

Ensemble

0.015440.133970.25250.37103May 8, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.05
0.033
2026.05
0.034
2026.05
0.048
2026.05
0.049
2026.05
0.051
2026.05
0.067
2026.05
0.067
2026.05
0.087
2026.05
0.087
2026.05
0.093
2026.05
0.093
2026.05
0.093
2026.05
0.093
2026.05
0.097
2026.05
0.199
2026.05
0.472