ResearchBenchmarksConversational Assistant on Preference Optimization ConversationalFollow0.28RewardZOPrO0.1240.16450.2050.2455Mar 5, 2025Evaluation ResultsMethodMethodLinksRewardZOPrOModel=Qwen 2.5, Parame...Model=Qwen 2.5, Parameters=1.5B2025.030.28ZOPrOModel=Gemma 2, Paramet...Model=Gemma 2, Parameters=2.4B2025.030.13