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

Cultural commonsense reasoning on CultureAtlas Low Resource

0.941Precision

GPT-4

Updated 4d ago

Evaluation Results

Method	Links
GPT-4 2026.01		0.941	0.901	0.921
CALM 2026.01		0.932	0.863	0.886
LLaMA-2 2026.01		0.87	0.207	0.335
Vicuna 2026.01		0.813	0.557	0.661
Vicuna 2026.01		0.678	0.823	0.743
LLaMA-2 2026.01		0.533	0.205	0.296