---
id: 20260513-T0-16
title: "MemQ：基于溯源DAG的自进化记忆体集成Q-Learning"
title_en: "MemQ: Self-Evolving Memory Agents with Q-Learning via Provenance DAGs"
url: https://ai.daily.yangsir.net/daily/20260513-T0-16
issue_date: 2026-05-13
publish_date: 2026-05-12T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2605.08374
---

# MemQ：基于溯源DAG的自进化记忆体集成Q-Learning

最新研究MemQ将Q-Learning集成到自进化记忆体代理中，解决当前LLM记忆体独立评估的问题。传统方法孤立评估记忆检索质量，忽略了记忆之间的依赖关系。MemQ通过溯源DAG结构，让记忆体能够持续优化检索策略，提升AI代理的长期学习效果。

## English Version

**MemQ: Self-Evolving Memory Agents with Q-Learning via Provenance DAGs**

MemQ integrates Q-learning into self-evolving memory agents over provenance DAGs, solving isolated memory evaluation in LLMs. Current methods assess retrieval quality without considering dependencies between memories. MemQ enables continuous optimization of retrieval strategies through DAG structure.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2605.08374)

**详情页**：https://ai.daily.yangsir.net/daily/20260513-T0-16

---

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