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

Mathematical Reasoning on GSM8K (test) (EM, ΔGSM8K)

66.26GSM8K EM

EPnG

46.281651.468356.65561.8417Jun 30, 2026Jul 1, 2026Jul 2, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.07
66.26-38.43
2026.07
65.13-38.06
2026.07
64.97-38.54
2026.07
64.44-36.7
2026.06
49.7--
2026.06
48.840.86-
2026.06
48.840.86-
2026.06
48.281.42-
2026.06
47.432.27-
2026.06
47.262.44-
2026.06
47.132.57-
2026.06
47.052.65-