让多个AI像人类一样分阶段讨论，达成可追溯的决策。
From Debate to Deliberation: Structured Collective Reasoning with Typed Epistemic Acts
- 设计四种角色和十四类思维动作，规范集体推理流程
- 在非例行任务中比自由辩论提升0.95分，隐藏信息任务达9.56分最优
- 输出带异议记录的结构化结论，适合需要过程透明的复杂决策
多智能体大模型系统日益用于复杂推理，但其交互模式仍局限于投票、无序辩论或流水线编排。现有方法均未实现「审议」：一种分阶段过程，参与者角色分化，交换类型化的认知行为，保留分歧，并收敛至可问责的结果。本文提出审议式集体智能（DCI），定义四种推理原型、14种类型化认知行为、共享工作空间及收敛流算法DCI-CF，确保终止并生成包含选定方案、残余异议、少数意见报告与重开条件的结构化决策包。在七个领域共45个任务上使用Gemini 2.5 Flash评估，非例行任务（n=40）中DCI显著优于无序辩论（+0.95，95%置信区间[+0.41, +1.54]）。在需整合视角的隐藏信息任务中表现最佳（9.56），而常规决策任务中表现较差（5.39），验证任务依赖性。DCI生成100%结构化决策包与98%少数意见报告，是所有基线缺失的成果。然而，其消耗约62倍单智能体令牌，且单智能体生成在整体质量上更优。核心贡献不在于更多智能体更好，而在于重大决策在过程可追溯的前提下，结构化审议带来价值。
原文摘要 · Abstract (English)
Multi-agent LLM systems increasingly tackle complex reasoning, yet their interaction patterns remain limited to voting, unstructured debate, or pipeline orchestration. None model deliberation: a phased process where differentiated participants exchange typed reasoning moves, preserve disagreements, and converge on accountable outcomes. We introduce Deliberative Collective Intelligence (DCI), specifying four reasoning archetypes, 14 typed epistemic acts, a shared workspace, and DCI-CF, a convergent flow algorithm that guarantees termination with a structured decision packet containing the selected option, residual objections, minority report, and reopen conditions. We evaluate on 45 tasks across seven domains using Gemini 2.5 Flash. On non-routine tasks (n=40), DCI significantly improves over unstructured debate (+0.95, 95% CI [+0.41, +1.54]). DCI excels on hidden-profile tasks requiring perspective integration (9.56, highest of any system on any domain) while failing on routine decisions (5.39), confirming task-dependence. DCI produces 100% structured decision packets and 98% minority reports, artifacts absent from all baselines. However, DCI consumes ~62x single-agent tokens, and single-agent generation outperforms DCI on overall quality. DCI's contribution is not that more agents are better, but that consequential decisions benefit from deliberative structure when process accountability justifies the cost.
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。