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

Zero-shot Reasoning on PIQA

82.54PIQA Zero-shot Accuracy

FP16

55.978462.874269.7776.6658Jul 25, 2023Jan 14, 2024Jul 6, 2024Dec 27, 2024Jun 19, 2025Dec 10, 2025Jun 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
82.54
2026.06
82.37
2026.06
82.32
2026.06
82.16
2026.06
82.15
2025.07
82.1
2025.07
81.77
2026.06
81.72
2026.06
81.39
2026.06
81.28
2023.07
80.9
2023.07
80.9
2025.07
80.71
2025.07
80.63
2023.07
80.52
2023.07
80.47
2026.06
80.36
2023.07
80.25
2025.07
80.19
2026.06
80.14
2025.07
79.87
2026.06
79.6
2026.06
79.05
2026.04
79
2025.07
78.67
2025.07
78.67
2026.02
78.6
2023.07
78.56
2026.06
78.07
2024.06
78.02
2025.07
77.97
2026.06
77.97
2026.06
77.91
2026.06
77.69
2025.07
77.62
2026.06
77.48
2026.06
77.37
2025.07
77.35
2025.07
77.31
2025.07
77.09
2026.06
77.04
2026.06
76.99
2026.06
76.55
2026.06
76.55
2026.02
76.4
2025.07
76.39
2026.06
76.28
2026.06
76.03
2025.07
75.84
2025.07
75.82
2024.06
75.41
2026.06
75.3
2025.07
75.14
2025.07
75.14
2026.06
75.12
2023.07
75.08
2026.06
74.16
2024.06
73.94
2025.07
73.69
2025.10
73.67
2025.10
73.12
2026.04
73
2026.04
73
2025.07
72.86
2025.10
72.57
2024.06
72.31
2025.10
72.25
2025.10
72.19
2026.04
72
2025.10
71.87
2024.06
71.81
2024.06
71.6
2025.10
71.32
2024.06
71.11
2025.07
71.01
2026.04
71
2025.10
70.94
2025.10
70.72
2025.10
70.45
2025.10
70.07
2025.10
70.02
2025.07
69.45
2024.06
68.44
2025.07
68.28
2026.04
68
2024.06
67.52
2026.02
67
2026.04
67
2026.04
66
2024.06
65.4
2026.02
65.2
2024.06
64.63
2026.04
64
2026.04
63
2024.06
61.8
2024.06
59.52
2024.06
59.52
2024.06
59.47
2026.02
57.3
2026.04
57
Showing 100 of 115 rows