Long-term conversational agents need memory systems that capture relationships between events, not merely isolated facts, to support temporal reasoning and multi-hop question answering. Current approaches face a fundamental trade-off: flat memory is efficient but fails to model relational structure, while graph-based memory enables structured reasoning at the cost of expensive and fragile construction. To address these issues, we propose \textbf{StructMem}, a structure-enriched hierarchical memory framework that preserves event-level bindings and induces cross-event connections. By temporally anchoring dual perspectives and performing periodic semantic consolidation, StructMem improves temporal reasoning and multi-hop performance on \texttt{LoCoMo}, while substantially reducing token usage, API calls, and runtime compared to prior memory systems, see https://github.com/zjunlp/LightMem .
@article{arxiv.2604.21748,
title = {StructMem: Structured Memory for Long-Horizon Behavior in LLMs},
author = {Buqiang Xu and Yijun Chen and Jizhan Fang and Ruobin Zhong and Yunzhi Yao and Yuqi Zhu and Lun Du and Shumin Deng},
journal= {arXiv preprint arXiv:2604.21748},
year = {2026}
}