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

Multilingual Mathematical Reasoning on MSVAMP (all)

89Accuracy

AdaMame-GRPO

66.1272.067883.94Jun 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
89860.9
2026.06
88.6851
2026.06
8884.31.2
2026.06
80.428.113.6
2026.06
7760.71.4
2026.06
76.619.715.2
2026.06
76.658.71.4
2026.06
7660.31.5
2026.06
74.621.715.2
2026.06
71.456.37.5
2026.06
70.454.47.3
2026.06
6727.913.1