---
title: "从原子动作到标准操作程序：面向自进化LLM代理的迭代工具优化"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.07321v1"
permalink: "https://feed.astralor.com/items/itm_1ff7cb29e11e18902b148a4d"
published: "2026-07-08T12:09:49.000Z"
score: 78
channel: "tech"
tags: ["EvoSOP","LLM Agent","工具优化"]
---

# 从原子动作到标准操作程序：面向自进化LLM代理的迭代工具优化

## AI 点评

> 这是一篇让代理自我进化的实用研究，尤其适合关注自主智能体和工具学习的开发者。它解决了静态工具集的核心痛点，并提供了完整的迭代框架。

## 速览

本研究提出EvoSOP框架，使大语言模型（LLM）代理能够将原子动作（如文件读写、单轮搜索）合成为可复用的标准操作程序（SOP），通过构建、合并、评估和剪枝的系统化生命周期实现工具集的迭代优化。实验证明，该方法显著提升任务成功率并减少交互轮数，为开发自进化代理提供了可扩展路径。

## 关键点

- 提出EvoSOP框架，让代理从执行轨迹中提取SOP作为高阶工具
- 包含构建、合并、评估和剪枝的完整工具优化生命周期
- 实验表明任务成功率显著提升，交互轮数大幅减少

## 全文

Tool utilization enables Large Language Model (LLM) agents to interact with the real world and resolve complex tasks. However, existing agent frameworks predominantly rely on static toolsets composed of granular atomic actions (e.g., basic file I/O or single-turn search), which forces agents to reinvent low-level logic for every recurring workflow, leading to increased reasoning overhead and failure rates. In this study, we propose that agents can achieve self-evolution by synthesizing these atomic actions into reusable Standard Operating Procedures (SOPs), which function as callable higher-order tools that encapsulate multi-step logic. We further introduce EvoSOP, a framework that empowers agents to extract SOPs from execution trajectories and iteratively optimize the toolset through a systematic lifecycle of construction, merging, evaluation, and pruning. Extensive experiments demonstrate that EvoSOP significantly boosts task success rates while substantially reducing the number of interaction rounds compared to baselines. Our analysis also reveals that iterative tool optimization fosters reliable and efficient tool-use patterns, providing a scalable pathway for the development of self-evolving agents.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_1ff7cb29e11e18902b148a4d