---
id: 20260409-T0-10
title: "EvolveRouter实现多代理问答的协同路由与提示"
title_en: "EvolveRouter Co-Evolves Routing and Prompts for Multi-Agent QA"
url: https://ai.daily.yangsir.net/daily/20260409-T0-10
issue_date: 2026-04-09
publish_date: 2026-04-08T04:00:00.000Z
source_name: "arXiv cs.CL (NLP)"
source_url: https://arxiv.org/abs/2604.05149
---

# EvolveRouter实现多代理问答的协同路由与提示

研究提出EvolveRouter方法，通过协同进化路由策略和提示优化多代理问答性能。现有路由方法通常在固定代理池中优化，而EvolveRouter动态调整代理组合和提示设计。该方法在复杂问答任务上比最佳基线提高25%，特别是在需要多个专家协作的场景中效果显著。这一突破为构建高效的多智能体系统提供了新思路，可应用于客服、教育等需要分工协作的领域。

## English Version

**EvolveRouter Co-Evolves Routing and Prompts for Multi-Agent QA**

Researchers present EvolveRouter, which co-evolves routing strategies and prompts to optimize multi-agent QA performance. Unlike methods optimizing over fixed pools, EvolveRouter dynamically adjusts agent combinations and prompt design. It improves performance by 25% over baselines on complex QA tasks, especially when multiple experts collaborate. This advances multi-agent system design for customer service and education.

---

**来源**：[arXiv cs.CL (NLP)](https://arxiv.org/abs/2604.05149)

**详情页**：https://ai.daily.yangsir.net/daily/20260409-T0-10

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*