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

Character-level Accuracy Reasoning on WinoGrande

64.01Character-level Accuracy

MobileLLM-Flash 1.4B

53.173255.986658.861.6134Mar 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
64.01
2026.03
61.09
2026.03
59.67
2026.03
59.35
2026.03
58.88
2026.03
58.8
2026.03
58.4
2026.03
58.2
2026.03
56.09
2026.03
55.96
2026.03
53.59