List your company

Critique Quality Evaluation on Math

57.5Win RateRCO
36.742.147.552.9Jun 27, 2025

Evaluation results

MethodLinks
RCOComparison Target=LLaMa-3-8B-Instruct + MSE + DPO2025.0657.522.52037.5
RCOComparison Target=LLaMa-3-8B-Instruct2025.0655301540
RCOComparison Target=LLaMa-3-70B-Instruct2025.0650104010
RCOComparison Target=LLaMa-3-8B-Instruct2025.0647.53517.530
RCOComparison Target=LLaMa-3-70B-Instruct2025.0642.532.52517.5
RCOComparison Target=LLaMa-3-8B-Instruct + MSE + DPO2025.0637.54517.520