Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Physical Commonsense Reasoning on PIQA (test)

90.7Accuracy

UL20B

73.90478.264582.62586.9855May 10, 2022Sep 2, 2022Dec 26, 2022Apr 20, 2023Aug 13, 2023Dec 6, 2023Mar 30, 2024
Updated 4d ago

Evaluation Results

MethodLinks
2022.05
90.7
2022.10
90.1
2022.05
90.1
2022.10
89.5
2022.10
87.41
2022.10
85.9
2024.03
84.66
2023.07
82.8
2023.07
82.8
2023.07
82.4
2023.07
82.3
2023.07
81.9
2023.07
81.9
2024.03
80.74
2023.07
80.6
2023.07
80.5
2023.07
80.1
2023.07
79.8
2022.10
79.4
2023.07
78.8
2024.03
78.07
2023.07
76.7
2024.03
75.14
2022.10
74.55