ResearchBenchmarksMachine Translation on Commonsense MTMachine Translation on Commonsense MTFollow4.2ScoreDUAL-REFLECT3.0563.3533.653.947Jun 11, 2024Evaluation resultsMethodMethodLinksScoreAccuracyDUAL-REFLECTBase Model=ChatGPTBase Model=ChatGPT2024.064.277.4GPT-42024.063.969.8ChatGPTStrategy=MAD (Multi-Ag...Strategy=MAD (Multi-Agent Debate)2024.063.776.2ChatGPTStrategy=Self-ReflectStrategy=Self-Reflect2024.063.464.9ChatGPTStrategy=RerankStrategy=Rerank2024.063.366.8ChatGPTPrompting Strategy=Zer...Prompting Strategy=Zero-Shot2024.063.163.8