arXiv：cs.CL· Zirui Liao, Zhengxian Wu, Zhuohong Chen, Yunyao Yu, Xiaoyu Liu, Yifan Xu, Haoqian Wang·· 3 小时前
CogMem：面向长期对话的可演化认知记忆架构
From Retrieval to Reconstruction: Constructing Evolvable Cognitive Memory for Long-Term Dialogue
AI 导读
研究者提出 CogMem 认知记忆架构，基于 PEC²F（Person-Event-Concept-Claim-Fact）图模式，用专门的 Claim 节点保留主观陈述的来源与对象，Fact 和 Event 节点分别表示语义与情景知识。
正文
Abstract:Large Language Models (LLMs) serving as long-term dialogue agents require memory systems that support reliable reasoning over extended interactions. However, existing Retrieval-Augmented Generation (RAG) frameworks typically treat memory as passive storage, making it difficult to distinguish source-attributed beliefs from unattributed event/fact records and to connect evidence dispersed across sessions. We introduce CogMem, a cognitive memory architecture based on the PEC$^2$F (Person-Event-Concept-Claim-Fact) graph schema. Dedicated Claim nodes preserve the source and target of subjective statements, while Fact and Event nodes represent semantic and episodic knowledge. Dialogue turns are incrementally converted into provenance-aware graph records, consolidated into higher-level facts, and reconciled into temporally scoped Claim views when the same source provides conflicting updates. For retrieval, a rule-based controller driven by LLM intent parsing composes four deterministic graph operators---anchoring, traversal, intersection, and evidence grounding---to reconstruct query-relevant context. Experiments on LoCoMo and LongMemEval show strong performance, especially on multi-hop, temporal, and knowledge-update tasks. Ablations and a semantic-collapse probe support complementary contributions from epistemic separation, consolidation, and agentic retrieval. Code: this https URL.
| Comments: | Accepted to EMNLP 2026 (main conference). 21 pages |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.11314 [cs.CL] |
| (or arXiv:2610.11314v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11314 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: ZiRui Liao [view email]
[v1]
Thu, 8 Oct 2026 06:21:45 UTC (1,417 KB)
来源：arXiv：cs.CL · arxiv.org