Membox: Weaving Topic Continuity into Long-Range Memory for LLM Agents

2026-01-07

arXiv.org

AI Summary

This work addresses the challenge that existing large language model agents struggle to maintain thematic continuity in dialogue, often resulting in fragmented narratives and broken causal chains. To overcome this, the authors propose Membox, a novel hierarchical memory architecture that introduces thematic continuity modeling at the storage stage. Membox employs Topic Loom to aggregate thematically related dialogue segments into coherent “memory boxes” and utilizes Trace Weaver to construct long-range event timelines across conversational discontinuities, enabling cognitively inspired, efficient memory organization. Departing from the conventional “fragmented storage–retrieval reconstruction” paradigm, Membox achieves up to a 68% relative improvement in temporal reasoning F1 score on the LoCoMo benchmark, significantly outperforming baselines such as Mem0 and A-MEM while using fewer context tokens, thereby achieving both higher efficiency and superior performance.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Modeling other Agents

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Human-agent dialogues often exhibit topic continuity-a stable thematic frame that evolves through temporally adjacent exchanges-yet most large language model (LLM) agent memory systems fail to preserve it. Existing designs follow a fragmentation-compensation paradigm: they first break dialogue streams into isolated utterances for storage, then attempt to restore coherence via embedding-based retrieval. This process irreversibly damages narrative and causal flow, while biasing retrieval towards lexical similarity. We introduce membox, a hierarchical memory architecture centered on a Topic Loom that continuously monitors dialogue in a sliding-window fashion, grouping consecutive same-topic turns into coherent"memory boxes"at storage time. Sealed boxes are then linked by a Trace Weaver into long-range event-timeline traces, recovering macro-topic recurrences across discontinuities. Experiments on LoCoMo demonstrate that Membox achieves up to 68% F1 improvement on temporal reasoning tasks, outperforming competitive baselines (e.g., Mem0, A-MEM). Notably, Membox attains these gains while using only a fraction of the context tokens required by existing methods, highlighting a superior balance between efficiency and effectiveness. By explicitly modeling topic continuity, Membox offers a cognitively motivated mechanism for enhancing both coherence and efficiency in LLM agents.

Problem

• topic continuity
• LLM agent memory
• dialogue coherence
• long-range memory
• temporal reasoning

Innovation

• topic continuity
• hierarchical memory architecture
• memory boxes
• long-range dialogue coherence
• LLM agent memory