提出路由增益认证框架，解决大模型多智能体系统中路由是否有效的判断难题。
RouteGuard: Certifying Routing Gain in LLM Multi-Agent Systems When Complementarity Is Not Enough
- 基于条件后悔函数构建可验证的路由增益评估机制
- 在11个跨模型基准上验证了路由增益依赖于特定工作负载单元
- 适用于需要可信部署决策的研发团队和系统集成者
多智能体大模型系统通过模型驱动的顾问进行路由，但部署前往往无法预知路由是否有效。现有路由器优化门控AUC，并假设顾问互补性足以保证收益。我们证明二者均无法决定可部署增益。本文提出RouteGuard，一个部署认证框架。路由增益分解为 $G = πΔ_E$，其可实现程度由条件后悔泛函 $Φ$ 决定，而非AUC。该框架提供有限样本认证区间，并具有匹配的Le Cam下界，在固定活动类上保持常数最优，且存在鲁棒性相变。在两个基准测试中，该框架发挥守卫作用：在RouterBench（11个跨家族模型）上，认证结果取决于采样单位——提示级采样下可认证优于GPT-4的增益，而工作负载簇重采样下则不认证，因增益仅依赖86个工作负载单元中的3个；在OpenRCA（三个Gemini顾问）中，顾问统计冗余：所有测试池（221个RouterBench池与3个OpenRCA分布）中实际最优解均位于或低于独立基线，协议正确拒绝认证。预先注册的半合成对照证实校准性：当 $m \ge m^\star$ 时，协议仅认证真实增益，不误认零效应。代码与冻结资产将在论文发布时公开。
原文摘要 · Abstract (English)
Multi-agent LLM systems route among model-backed advisors, yet a deployer rarely knows before shipping whether routing will help at all. Prevailing routers optimize a gate's AUC and presume that advisor complementarity suffices. We show that neither determines the deployable gain. We introduce RouteGuard, a deployment-certification framework. Routing gain decomposes as $G = πΔ_E$, and the achievable gain is governed by a conditional-regret functional $Φ$, not by AUC. A finite-sample certification bracket comes with a matching Le Cam lower bound, constant-sharp over the fixed-activity class, and a robustness phase transition. On two benchmarks the framework acts as a guardrail. On RouterBench (11 cross-family models) the verdict depends on the sampling unit: the protocol certifies a gain over GPT-4 under prompt-level sampling and withholds it under workload-cluster resampling, because the gain rests on 3 of 86 workload cells. On OpenRCA (three Gemini advisors) the advisors are statistically redundant: the realized oracle sits at or below the independence baseline in all pools we tested (221 RouterBench pools and three OpenRCA distributions), so the protocol correctly refuses to certify. A pre-registered semi-synthetic control confirms calibration: the protocol certifies a genuine gain once $m \ge m^\star$ and does not certify a true null. Code and frozen artifacts will be released with the published version.
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。