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

Mathematics Question Answering on GSM8K (test, zero-shot)

80.67Flexible EM

Original

1.744422.234742.72563.2153Feb 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
80.6780.52
2026.02
76.1375.92
2026.02
75.8475.56
2026.02
75.5375.26
2026.02
75.4275.27
2026.02
75.3675.04
2026.02
75.2875.11
2026.02
75.1274.37
2026.02
74.1573.92
2026.02
74.1573.46
2026.02
38.2138.13
2026.02
33.2533.02
2026.02
32.9232.57
2026.02
32.8132.75
2026.02
32.6232.27
2026.02
32.3332.01
2026.02
32.1632.04
2026.02
32.0431.75
2026.02
31.9931.54
2026.02
28.1826.84
2026.02
22.9722.97
2026.02
18.4218.27
2026.02
18.1218.04
2026.02
18.0417.82
2026.02
18.0317.03
2026.02
17.8917.82
2026.02
17.5917.51
2026.02
17.3617.29
2026.02
17.1317.06
2026.02
15.2411.45
2026.02
9.861.44
2026.02
4.784.62