推理分歧之处：本地化多智能体辩论
Where Reasoning Diverges: Localized Multi-Agent Debate for Multi-Hop Question Answering
浏览论文内容
中文总结 AI 辅助
针对多智能体辩论冗余交换完整推理轨迹的问题，提出LMAD协议，通过定位冲突并限制辩论范围，在十个骨干模型上实现宏平均评判准确率提升7.20个百分点。
中文摘要 AI 辅助
多智能体辩论通常会交换完整的推理轨迹，即便分歧仅涉及少数中间主张。我们提出本地化多智能体辩论（Localized Multi-Agent Debate, LMAD），这是一种推理时协议，它将智能体轨迹表示为带类型的节点，定位其最早冲突并将辩论限制在相应的局部片段中。受保护的解决方案扩展了共享的已提交状态，以便后续冲突无需重新打开已接受的步骤即可解决。我们在四个多跳问答基准上使用来自四个模型家族的十个骨干模型评估单一固定的LMAD配置。我们的方法在所有十个骨干模型上实现了最高的宏平均评判准确率，比最强的常规基准高出最多7.20个百分点。
英文摘要
Multi-agent debate commonly exchanges complete rationales even when disagreements concern only a few intermediate claims. We introduce Localized Multi-Agent Debate (LMAD), an inference-time protocol that represents agent rationales as nodes, locates their earliest conflict, and restricts debate to the corresponding local segments. Guarded resolution extends a shared committed state so that later conflicts can be addressed without reopening accepted steps. We evaluate LMAD on four multi-hop question-answering benchmarks using ten backbones from four model families. Our method achieves the highest macro-averaged judge accuracy across all ten backbones, outperforming the strongest conventional baseline by up to 7.20 percentage points.
发表机构
- The Chinese University of Hong Kong（香港中文大学）
- Nagoya University（名古屋大学）
- Institute of Science Tokyo（东京科学大学）
- University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
机构由 AI 辅助整理，请以论文原文为准。
