---
id: 20260321-T0-17
title: "检索增强LLM代理：从经验中学习通用任务能力"
title_en: "Retrieval-Augmented LLM Agents: Learning from Experience for General Tasks"
url: https://ai.daily.yangsir.net/daily/20260321-T0-17
issue_date: 2026-03-21
publish_date: 2026-03-20T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2603.18272
---

# 检索增强LLM代理：从经验中学习通用任务能力

新研究提出检索增强LLM代理框架，解决泛化到未见任务的问题。当前方法依赖微调或无训练记忆，泛化能力有限。该方案结合检索增强与经验学习，使代理能够从历史交互中提取模式，提升对新任务的适应能力。实验显示，该框架在多样化任务中表现优于传统方法，为通用智能代理发展提供新路径。

## English Version

**Retrieval-Augmented LLM Agents: Learning from Experience for General Tasks**

This research introduces retrieval-augmented LLM agents to solve generalization challenges to unseen tasks. Current approaches rely on either fine-tuning or training-free memory, which have limited generalization. This framework combines retrieval enhancement with experience learning, enabling agents to extract patterns from historical interactions and improve adaptability to new tasks. Experiments show superior performance over traditional methods across diverse tasks.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2603.18272)

**详情页**：https://ai.daily.yangsir.net/daily/20260321-T0-17

---

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