Modern large language models (LLMs) like ChatGPT have shown remarkable performance on general language tasks but still struggle on complex reasoning tasks, which drives the research on cognitive behaviors of LLMs to explore human-like problem-solving strategies. Along this direction, one representative strategy is self-reflection, which asks an LLM to refine the solution with the feedback generated by itself iteratively. However, our study shows that such reflection-style methods suffer from the Degeneration-of-Thought (DoT) problem: once the LLM has established confidence in its solutions, it is unable to generate novel thoughts later through reflection even if its initial stance is incorrect. To address the DoT problem, we propose a Multi-Agent Debate (MAD) framework, in which multiple agents express their arguments in the state of "tit for tat" and a judge manages the debate process to obtain a final solution. Clearly, our MAD framework encourages divergent thinking in LLMs which would be helpful for tasks that require deep levels of contemplation. Experiment results on two challenging datasets, commonsense machine translation and counter-intuitive arithmetic reasoning, demonstrate the effectiveness of our MAD framework. Extensive analyses suggest that the adaptive break of debate and the modest level of "tit for tat" state are required for MAD to obtain good performance. Moreover, we find that LLMs might not be a fair judge if different LLMs are used for agents. Codes: https://github.com/Skytliang/Multi-Agents-Debate

翻译：现代大规模语言模型（如ChatGPT）在通用语言任务上展现出卓越性能，但在复杂推理任务上仍存在困难，这推动了对大语言模型认知行为的研究，以探索类人问题解决策略。沿此方向，一种代表性策略是自我反思，即让大语言模型利用自身生成的反馈反复改进解决方案。然而，我们的研究表明此类反思方法存在“思维退化”（Degeneration-of-Thought，DoT）问题：一旦大语言模型对其解决方案建立信心，即便初始立场错误，后续反思也无法生成新思路。为解决DoT问题，我们提出多智能体辩论框架（Multi-Agent Debate，MAD），其中多个智能体以“针锋相对”状态表达论点，并由裁判管理辩论过程以获得最终解决方案。显然，我们的MAD框架促进了大语言模型中的发散思维，这对需要深度思考的任务具有助益。在常识机器翻译和反直觉算术推理两个具有挑战性的数据集上的实验结果表明了MAD框架的有效性。广泛分析表明，为获得良好性能，MAD需要自适应终止辩论并保持适度程度的“针锋相对”状态。此外，我们发现若不同智能体使用不同大语言模型，大语言模型可能无法成为公平的裁判。代码：https://github.com/Skytliang/Multi-Agents-Debate