TraceMem: Weaving Narrative Memory Schemata from User Conversational Traces

2026-02-10

AI Summary

This work addresses the challenge of maintaining narrative coherence in long-term dialogues, a limitation imposed by the fixed context windows of large language models. Inspired by cognitive science, the authors propose a three-stage framework that integrates deductive topic segmentation with two-phase memory consolidation—synaptic and systems-level—to construct structured, temporally evolving narrative memory schemas. These schemas are further enhanced through hierarchical clustering and an agent-driven retrieval mechanism to support complex reasoning. Notably, this approach is the first to explicitly incorporate narrative coherence into dialogue memory modeling, yielding thematically unified and temporally progressive memory traces. Evaluated on the LoCoMo benchmark, the method achieves state-of-the-art performance, significantly outperforming existing approaches in multi-hop and temporal reasoning tasks.

Technology Category

• Natural Language Processing: Conversational AI/Dialog Systems
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Sustaining long-term interactions remains a bottleneck for Large Language Models (LLMs), as their limited context windows struggle to manage dialogue histories that extend over time. Existing memory systems often treat interactions as disjointed snippets, failing to capture the underlying narrative coherence of the dialogue stream. We propose TraceMem, a cognitively-inspired framework that weaves structured, narrative memory schemata from user conversational traces through a three-stage pipeline: (1) Short-term Memory Processing, which employs a deductive topic segmentation approach to demarcate episode boundaries and extract semantic representation; (2) Synaptic Memory Consolidation, a process that summarizes episodes into episodic memories before distilling them alongside semantics into user-specific traces; and (3) Systems Memory Consolidation, which utilizes two-stage hierarchical clustering to organize these traces into coherent, time-evolving narrative threads under unifying themes. These threads are encapsulated into structured user memory cards, forming narrative memory schemata. For memory utilization, we provide an agentic search mechanism to enhance reasoning process. Evaluation on the LoCoMo benchmark shows that TraceMem achieves state-of-the-art performance with a brain-inspired architecture. Analysis shows that by constructing coherent narratives, it surpasses baselines in multi-hop and temporal reasoning, underscoring its essential role in deep narrative comprehension. Additionally, we provide an open discussion on memory systems, offering our perspectives and future outlook on the field. Our code implementation is available at: https://github.com/YimingShu-teay/TraceMem

Problem

• long-term interaction
• narrative coherence
• conversational memory
• context window
• dialogue history

Innovation

• narrative memory schemata
• conversational traces
• memory consolidation
• hierarchical clustering
• long-term dialogue