List your company

Multi-task Language Understanding on MMLU (0-shot)

69.11Exact Match (EM)VAR
30.983640.881850.7860.6782Feb 16, 2025

Evaluation results

MethodLinks
VARBase Model=Qwen2.5-7B2025.0269.11
DPOBase Model=Qwen2.5-7B2025.0268.64
ALoLBase Model=Qwen2.5-7B2025.0268.62
BaseBase Model=Qwen2.5-7B2025.0267.13
VARBase Model=Llama2-7B2025.0238.57
BaseBase Model=Llama2-7B2025.0237.46
ALoLBase Model=Llama2-7B2025.0235.78
DPOBase Model=Llama2-7B2025.0232.45