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

Mathematical Reasoning on GSM8K (Accuracy by Sample Count)

94.39Task Accuracy

DynaSteer

51.136462.365773.59584.8243May 19, 2026May 25, 2026May 31, 2026Jun 7, 2026Jun 13, 2026Jun 19, 2026Jun 26, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.06
94.39-----
2026.06
77.94-----
2026.06
76.95-----
2026.05
57706767.264.961.7
2026.05
55.168.965.164.664.961.7
2026.05
54.569.364.166.366.459.6
2026.05
54.267.963.564.661.153.2
2026.05
5466.862.463.766.461.7
2026.05
52.866.662.565.665.657.4