发现多智能体辩论中投票比辩论本身更有效，简单集成已足够好。
Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?
- 拆解辩论为投票与对话两部分，分别评估贡献。
- 投票单独就能带来多数性能提升，辩论不改期望正确率。
- 提出理论模型解释为何辩论无效，适合想优化推理的工程师。
多智能体辩论（MAD）作为一种通过协作推理提升大语言模型性能的新兴范式，近年来备受关注。然而其有效性关键因素仍不明确。本文将MAD分解为多数投票与智能体间辩论两个核心组件，并在七个NLP基准上进行广泛实验。结果表明，多数投票本身已贡献了绝大多数性能提升，通常归因于辩论的效果。为此，我们提出一个理论框架，将辩论建模为随机过程，证明其在智能体信念轨迹上诱导出鞅（martingale），意味着辩论无法提高期望正确率。基于此，我们进一步验证：通过引导信念更新向修正方向偏移，可显著提升辩论效果。总体而言，尽管MAD有潜力，但在多数实际场景中，简单的集成方法仍是更强大且可靠的替代方案。代码已开源：https://github.com/deeplearning-wisc/debate-or-vote。
原文摘要 · Abstract (English)
Multi-Agent Debate~(MAD) has emerged as a promising paradigm for improving the performance of large language models through collaborative reasoning. Despite recent advances, the key factors driving MAD's effectiveness remain unclear. In this work, we disentangle MAD into two key components--Majority Voting and inter-agent Debate--and assess their respective contributions. Through extensive experiments across seven NLP benchmarks, we find that Majority Voting alone accounts for most of the performance gains typically attributed to MAD. To explain this, we propose a theoretical framework that models debate as a stochastic process. We prove that it induces a martingale over agents' belief trajectories, implying that debate alone does not improve expected correctness. Guided by these insights, we demonstrate that targeted interventions, by biasing the belief update toward correction, can meaningfully enhance debate effectiveness. Overall, our findings suggest that while MAD has potential, simple ensembling methods remain strong and more reliable alternatives in many practical settings. Code is released in https://github.com/deeplearning-wisc/debate-or-vote.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。