arXiv 2605.099422026-05-11cs.AI

HAGE: 通过RL驱动的加权图进化利用代理记忆

HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution

Department of Computer Science, The University of Texas at Dallas（德克萨斯大学达拉斯分校计算机科学系）；Department of Electrical and Computer Engineering, University of Florida（佛罗里达大学电气与计算机工程系）；University of California, Davis（加州大学戴维斯分校）

Dongming Jiang, Yi Li, Guanpeng Li, Qiannan Li, Bingzhe Li

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

HAGE提出了一种加权多关系记忆框架，通过RL驱动的训练框架优化路由行为和边表示，提升长周期推理准确性和效率。

摘要：

在代理大语言模型系统中，记忆检索常被视为静态查找问题，依赖于扁平向量搜索或固定二进制关系图。然而，固定图结构无法捕捉事件间关系的变动强度、置信度和查询依赖的相关性。本文提出HAGE，一种加权多关系记忆框架，将检索重新概念化为在统一关系记忆图上的序列、查询条件下的遍历。记忆被组织为关系特定的图视图，覆盖共享内存节点，每条边关联有可训练的关系特征向量编码多种关系信号。给定查询，基于LLM的分类器识别关系意图，路由网络动态调节对应边嵌入的维度。遍历分数通过学习的语义相似性和这些查询条件下的边表示组合计算。这使记忆遍历能够优先选择高效用的关系路径，同时软抑制噪声或弱相关连接。除了自适应遍历，HAGE进一步引入基于强化学习的训练框架，利用下游任务联合优化路由行为和边表示。最终，实验证明相比最先进的代理记忆系统，HAGE在长周期推理准确性和准确性-效率权衡上有所提升。我们的代码可在https://github.com/FredJiang0324/HAGE_MVPReview获取。

英文摘要：

Memory retrieval in agentic large language model (LLM) systems is often treated as a static lookup problem, relying on flat vector search or fixed binary relational graphs. However, fixed graph structures cannot capture the varying strength, confidence, and query-dependent relevance of relationships between events. In this paper, we propose HAGE, a weighted multi-relational memory framework that reconceptualizes retrieval as sequential, query-conditioned traversal over a unified relational memory graph. Memory is organized as relation-specific graph views over shared memory nodes, where each edge is associated with a trainable relation feature vector encoding multiple relational signals. Given a query, an LLM-based classifier identifies the relational intent, and a routing network dynamically modulates the corresponding dimensions of the edge embedding. Traversal scores are computed via a learned combination of semantic similarity and these query-conditioned edge representations. This allows memory traversal to prioritize high-utility relational paths while softly suppressing noisy or weakly relevant connections. Beyond adaptive traversal, HAGE further introduces a reinforcement learning-based training framework that jointly optimizes routing behavior and edge representations using downstream tasks. Finally, empirical results demonstrate improved long-horizon reasoning accuracy and a favorable accuracy-efficiency trade-off compared to state-of-the-art agentic memory systems. Our code is available at https://github.com/FredJiang0324/HAGE_MVPReview.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。