StructMem: Structured Memory for Long-Horizon Behavior in LLMs

arXiv:2604.2174892.44 citationsHas Code
AI Analysis

For developers of long-term conversational agents, StructMem offers a more efficient memory system that balances relational structure with computational cost.

StructMem introduces a hierarchical memory framework that preserves event-level bindings and induces cross-event connections, improving temporal reasoning and multi-hop QA on LoCoMo while reducing token usage, API calls, and runtime.

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 .

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes