List your company

Creative Writing on Creative Writing Alpaca-Eval 100 problems 2.0

93.81Length-Controlled Win RateICRL prompting
58.106867.375976.64585.9141May 21, 2025

Evaluation results

MethodLinks
ICRL promptingComparison baseline=Best-of-N2025.0593.811.01
ICRL promptingComparison baseline=Self-Refine2025.0586.323.03
ICRL promptingComparison baseline=Long CoT2025.0578.361.99
ICRL promptingComparison baseline=Reflexion2025.0559.483.47