List your company

Critique Quality Evaluation on Human Evaluation Overall

66Win RateRCO
36.8844.445259.56Jun 27, 2025

Evaluation results

MethodLinks
RCOComparison Target=LLaMa-3-70B-Instruct2025.066618.515.550.5
RCOComparison Target=LLaMa-3-8B-Instruct2025.064924.526.522.5
RCOComparison Target=LLaMa-3-8B-Instruct2025.0648351731
RCOComparison Target=LLaMa-3-8B-Instruct + MSE + DPO2025.0647.525.52720.5
RCOComparison Target=LLaMa-3-70B-Instruct2025.064333.523.519.5
RCOComparison Target=LLaMa-3-8B-Instruct + MSE + DPO2025.063832.529.58.5