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

Multiple-choice Question Answering on MMLU-Pro (Calibration and Ranking)

0.037Brier Score

GUPO + SAGE

0.022120.122560.2230.32344Jun 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
0.0370.1130.5720.4120.5120.57
2026.06
0.0450.120.1950.3470.3880.38
2026.06
0.0510.1440.2610.4030.4380.46
2026.06
0.0680.205-0.010.1630.1670.18
2026.06
0.0940.1790.10.3190.3420.355
2026.06
0.1130.2220.1240.3560.4120.445
2026.06
0.1260.2460.0690.3030.3120.325
2026.06
0.2140.3160.090.1880.1830.18
2026.06
0.4090.590.2990.2910.2960.34