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

Physical Commonsense Reasoning on PIQA (Accuracy and Normalized Accuracy)

82.54Accuracy

BF16

46.826456.098265.3774.6418Apr 26, 2026May 6, 2026May 16, 2026May 26, 2026Jun 5, 2026Jun 15, 2026Jun 25, 2026
Updated 26d ago

Evaluation Results

MethodLinks
2026.05
82.54-
2026.05
82.26-
2026.05
81.45-
2026.05
80.52-
2026.05
80.36-
2026.05
79.87-
2026.05
79.82-
2026.05
79.6-
2026.05
79.49-
2026.05
79.38-
2026.05
79-
2026.05
78.73-
2026.05
78.56-
2026.05
78.56-
2026.05
78.51-
2026.05
78.13-
2026.05
78.02-
2026.05
77.91-
2026.05
77.8-
2026.05
77.31-
2026.05
77.09-
2026.05
76.93-
2026.05
76.33-
2026.05
75.46-
2026.05
75.19-
2026.05
74.97-
2026.05
74.54-
2026.05
74.48-
2026.05
74.37-
2026.05
74.32-
2026.05
74.27-
2026.05
73.94-
2026.05
73.88-
2026.05
73.72-
2026.05
73.56-
2026.05
72.36-
2026.05
71.71-
2026.06
71.3-
2026.06
70.2-
2026.06
68.2-
2026.06
67.7-
2026.05
65.94-
2026.06
62.8-
2026.06
62.5-
2026.04
61.5-
2026.04
61.3-
2026.04
59.4759.58
2026.04
57.7857.89
2026.04
50.8-
2026.04
49.5-
2026.04
48.2-