发表机构
Zhejiang University; University of Electronic Science and Technology of China; University of Science and Technology of China（浙江大学; 电子科技大学; 中国科学技术大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
SEPAL通过三个私有Actor-Critic团队分别负责推理、证据和验证，并在答案级融合投票，提升了LLM多智能体协作的准确率。
AI 中文摘要
多智能体协作使大型语言模型（LLMs）能够通过深思和反馈改进问答。然而，共享讨论将修正与暴露于相同错误耦合在一起，这可能削弱投票所需的多样性。自一致性提供无反馈的采样多样性，而单对Actor-Critic协作仅优化一个候选。我们提出SEPAL，它分配三个私有Actor-Critic团队分别负责直接推理、证据基础和验证。角色特定训练赋予团队超越采样变化的推理目标。每个Critic在其团队内指导修订，防止反馈在候选之间传播错误。修订结束后，多数投票仅结合最终答案，保持推理历史分离直至决策。在五个开放权重骨干和五个问答基准上，SEPAL相对于匹配的单Actor-Critic对将平均准确率提高了1.81个百分点，且在所有五个骨干上均有改进。代码可在该URL获取。
英文摘要
Multi-agent collaboration lets large language models (LLMs) improve question answering through deliberation and feedback. Yet shared discussion couples correction with exposure to the same mistakes, which can erode the diversity needed for voting. Self-consistency offers sampling diversity without feedback, while single-pair Actor-Critic collaboration refines only one candidate. We introduce SEPAL, which assigns three private Actor-Critic teams to direct reasoning, evidence grounding, and verification. Role-specific training gives the teams different reasoning objectives beyond sampling variation. Each Critic guides revisions within its own team, preventing feedback from carrying errors across candidates. Once revision ends, majority voting combines only the final answers, keeping the reasoning histories separate until the decision. Across five open-weight backbones and five question-answering benchmarks, SEPAL improves mean accuracy by 1.81 percentage points over a matched single Actor-Critic pair, with improvements across all five backbones. Code is available at https://github.com/zhansan114514/SEPAL.
Comments22 pages, 4 figures. Code: https://github.com/zhansan114514/SEPAL
