arXiv：cs.CL· Lingxi Cui, Jinhao Ruan, Wenqing Wei, Guo Yi, Li Rao, Ke Chen, Lidan Shou·· 4 小时前
EGCForest：面向企业群聊的结构化智能体记忆组织框架
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats
AI 导读
EGCForest 是一个面向企业群聊的多视图分层记忆框架，将不可变消息日志组织为 Channel、Actor、Artifact 三片森林，并通过 Placement Map 与稀疏结构桥实现跨视图连接。在三个企业群聊基准上，该框架持续优于 RAG 与记忆系统基线，答案准确率较 MemOS 最高提升 20.85 分，并支持高效在线检索。
正文
Abstract:Enterprise group chats record project progress, decisions, task ownership, and other organizational information valuable to enterprise agents. Reliable retrieval is challenging because evidence spans interleaved threads, channels, and time, with meaning and authority conditioned on the speaker. Neither RAG's independent chunk retrieval nor memory systems' summary-based consolidation can reliably recover such distributed evidence. We present EGCForest, a novel multi-view hierarchical memory framework for enterprise group chats. EGCForest organizes an immutable message log into three complementary forests: a Channel forest for temporal workflows, an Actor forest for speaker-conditioned context, and an Artifact forest for deliverable lifecycles. Cross-view connections are enabled by a Placement Map and sparse structural bridges. Built on this memory backbone, EGCForest retrieves and assembles evidence by jointly leveraging message-level relevance, hierarchical context from forest summaries, and cross-view links, recovering distributed evidence. Experiments on three enterprise-chat benchmarks show that EGCForest consistently outperforms RAG and memory-system baselines, improving answer accuracy by up to 20.85 points over MemOS while supporting efficient online retrieval.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.19836 [cs.CL] |
| (or arXiv:2610.19836v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19836 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lingxi Cui [view email]
[v1]
Wed, 14 Oct 2026 09:07:53 UTC (508 KB)
来源：arXiv：cs.CL · arxiv.org