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

Physical Commonsense Reasoning on PIQA (Character-level Accuracy)

75.52Character-level Accuracy

MobileLLM-Flash 1.4B

68.052869.991471.9373.8686Mar 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
75.52
2026.03
75.41
2026.03
75.14
2026.03
74.27
2026.03
73.8
2026.03
71.82
2026.03
71.11
2026.03
70.08
2026.03
69.86
2026.03
69.48
2026.03
68.34