Files
myWiki/raw/papers/yao-ace-router-2026.md

52 lines
2.0 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

---
title: "ACE-Router: Generalizing History-Aware Routing from MCP Tools to the Agent Web"
created: 2026-06-19
updated: 2026-06-19
type: paper-raw
source: https://arxiv.org/abs/2601.08276
arxiv_id: 2601.08276
version: v2
---
# ACE-Router: Generalizing History-Aware Routing from MCP Tools to the Agent Web
**Authors**: Zhiyuan Yao (ZJU), Zishan Xu (SJTU), Yifu Guo (SYSU), Zhiguang Han (NTU), Cheng Yang (HDU), Shuo Zhang, Weinan Zhang (SJTU), Xingshan Zeng, Weiwen Liu (Huawei)
**Published**: 2026-01-13 (v2: 2026-04-19)
**Venue**: arXiv:2601.08276 (cs.AI)
**Code**: https://github.com/euyis1019/ACE-Router
## 核心洞察
ACE-Router 将 MCP 工具选择重新定义为**训练一个历史感知路由器**的问题——不是用 embedding 做静态匹配,而是让路由器理解多轮对话历史来做上下文感知的精确路由。
## 三大阶段
### 1. Candidate Graph + Self-Evolutionary Mutation
- 基于语义相似度构建候选图(阈值 τ=0.82
- 五种变异算子Function Enhancement, Parameter Mutation, Workflow Chaining, Helper Operation, Usage Extension
- 627 初始工具 → 2005 工具(通过变异扩展)
### 2. Trajectory Synthesis多 Agent 模拟)
- 从候选图采样(随机游走 DFS
- Planner Agent + User Agent + Assistant Agent + Tool Agent 四角色模拟
- 环境无关设计:无需真实 APILLM 模拟执行结果
- 产出 15,092 个历史感知路由训练样本
### 3. Light Routing Agent (LRA)
- 仅两个工具router_invoke + tool_execute
- 解耦路由决策与任务执行
- 可插拔:适配工具路由和 Agent 路由
## 关键结果
| 方法 | MCP-Universe | MCP-Mark |
|------|:---:|:---:|
| Text-Emb-3-Large (Q) | ~40.95% | ~29.89% |
| ReAct (Gemini-2.5-Pro) | ~41.80% | ~50.00% |
| GPT-4o Router | ~47.41% | ~48.00% |
| **ACE-Router (Qwen3-8B)** | **53.44%** | **60.00%** |
- 扩展候选池ReAct 41.80→36.47%ACE-Router 稳定在 53.02%
- 噪声环境GPT-4o 28% / Gemini 32%ACE-Router 保持 56%
- 多 Agent 泛化无需额外训练router 直接泛化到 Agent 路由