List your company

Logical Reasoning on BBH 3-shot chain-of-thought

61.35EMVAR
10.826823.943437.0650.1766Feb 16, 2025

Evaluation results

MethodLinks
VARBase Model=Qwen2.5-7B2025.0261.35
ALoLBase Model=Qwen2.5-7B2025.0261.14
DPOBase Model=Qwen2.5-7B2025.0253.06
DPOBase Model=Llama2-7B2025.0239.1
ALoLBase Model=Llama2-7B2025.0238.16
VARBase Model=Llama2-7B2025.0237.56
BaseBase Model=Qwen2.5-7B2025.0229.3
BaseBase Model=Llama2-7B2025.0212.77