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

Common Sense on OpenBookQA

81.8Accuracy

Temp-Scale

71.19273.94676.779.454Mar 10, 2026
Updated 4mo ago

Evaluation Results

MethodLinks
2026.03
81.81.1210.222.4
2026.03
81.80.6822.210.6
2026.03
80.80.8714.640.2
2026.03
80.41.3712.727.3
2026.03
80.20.6941.49.8
2026.03
80.21.1216.832.1
2026.03
800.4475.414.6
2026.03
79.80.9826.415.2
2026.03
79.80.431626.2
2026.03
79.60.7242.812.4
2026.03
79.40.72910.330.2
2026.03
78.80.4766.717.4
2026.03
78.60.6651428.8
2026.03
78.20.9129.814.6
2026.03
74.61.3825.247.2
2026.03
74.20.6542.629.3
2026.03
740.6521.515.2
2026.03
73.60.6524.129
2026.03
73.60.6675.229.3
2026.03
73.40.653.729.8
2026.03
71.60.77310.720.1