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

Commonsense Reasoning on HellaSWAG (Normalized Accuracy)

63.84HellaSWAG Normalized Accuracy

Gated KalmaNet

48.229652.282356.33560.3877Nov 26, 2025
Updated 2mo ago

Evaluation Results

MethodLinks
2025.11
63.84
2025.11
62.8
2025.11
62.23
2025.11
61.13
2025.11
60.96
2025.11
48.83