DIANOIA: 多智能体推理的诊断性分解与联合优化
DIANOIA: Diagnostic Decomposition and Joint Optimization for Multi-Agent Reasoning
AI总结：
提出DIANOIA框架，通过覆盖度、保真度和综合度三个可测量通道分解多智能体推理增益，并基于此设计诊断协议和对应系统，在多个基准上以更少token实现更优性能。
摘要：
多智能体LLM系统持续优于单智能体基线，但从业者仍无法预测哪种设计适用于新任务或诊断失败原因。我们认为这一差距主要源于该领域缺乏具有可测量原语和可测试预测的诊断框架。我们引入 extbf{DIANOIA}，将多智能体推理增益分解为覆盖度、保真度和综合度三个通道，每个通道均可经验测量。基于此分解，我们推导出一个诊断协议，可识别任何给定任务的瓶颈通道。我们将该协议实例化为一个多智能体系统，其三个组件与通道对应：角色多样化的提议者（覆盖度）、基于执行验证的验证者（保真度）和迭代综合者。在GSM8K、AIME-2025、MBPP和BFCL-SP上，我们的方法在匹配token预算下优于强多智能体基线，在MBPP上以约$5 imes$的token节省主导帕累托前沿，在匹配成本下达到$+4.6$pp。在每个基准上，协议都能正确选择瓶颈通道；我们围绕它构建的系统在多个模型上领先。我们发布代码、适配器、诊断指标和Claude Code技能，网址为https://anonymous.4open.science/r/DIANOIA4MAS。DIANOIA将多智能体设计重新定义为通道感知的资源分配：诊断你的任务的瓶颈通道，然后相应投入token。
英文摘要：
Multi-agent LLM systems consistently outperform single-agent baselines, yet practitioners still cannot predict which design works for a new task or diagnose why one fails. We argue this gap persists largely because the field lacks a diagnostic framework with measurable primitives and testable predictions. We introduce \textbf{DIANOIA}, a three-channel decomposition of multi-agent reasoning gain into coverage, fidelity, and synthesis, each of which is empirically measurable. From this decomposition, we derive a diagnostic protocol that identifies the bottleneck channels for any given task. We instantiate the protocol as a multi-agent system whose three components mirror the channels: role-diverse proposers for coverage, execution-grounded verification for fidelity, and iterative synthesis. On GSM8K, AIME-2025, MBPP, and BFCL-SP, our method outperforms strong multi-agent baselines under matched token budgets, dominating the Pareto frontier on MBPP at $\sim$$5{\times}$ token savings and reaching $+4.6$pp at matched cost. On every benchmark, the protocol picks the right bottleneck channels; the system we built around it leads across models. We release code, adapters, diagnostic metrics, and a Claude Code skill at https://anonymous.4open.science/r/DIANOIA4MAS. DIANOIA reframes multi-agent design as channel-aware resource allocation: diagnose which channel is the bottleneck for your task, then invest tokens accordingly.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
