---
id: 20260613-T0-02
title: "Evoflux：工具工作流的进化式推理，降低智能体部署成本"
title_en: "Evoflux Evolves Tool Workflows to Compact Agent Deployment"
url: https://ai.daily.yangsir.net/daily/20260613-T0-02
issue_date: 2026-06-13
publish_date: 2026-06-12T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2606.12674
---

# Evoflux：工具工作流的进化式推理，降低智能体部署成本

MIT团队提出Evoflux方法，通过进化算法实时优化工具工作流，使紧凑语言模型（LMs）能高效调用复杂工具。现有MCP工具调用存在依赖维护难题，而Evoflux能在推理时动态调整工具组合，减少90%的冗余调用。实验显示，该方法在电商推荐、代码生成等场景中，将响应速度提升3倍，部署成本降低60%。

## English Version

**Evoflux Evolves Tool Workflows to Compact Agent Deployment**

MIT researchers developed Evoflux, an evolutionary method that optimizes tool workflows in real-time for compact language models. It solves dependency maintenance issues in MCP-style tool calling by dynamically adjusting tool combinations, reducing redundant calls by 90%. Tests show 3x faster response times and 60% lower deployment costs in e-commerce and code generation scenarios.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2606.12674)

**详情页**：https://ai.daily.yangsir.net/daily/20260613-T0-02

---

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