Beyond Dialogue Time: Temporal Semantic Memory for Personalized LLM Agents

2026-01-12

AI Summary

This work addresses the limitations of existing memory mechanisms in large language model agents, which typically organize memories based on dialogue timestamps, leading to inaccurate temporal modeling and fragmented memory representations that struggle to capture persistent states and their evolution. To overcome this, the paper proposes the Temporal Semantic Memory (TSM) framework, which introduces a semantic timeline in place of the conventional chronological one. TSM constructs coherent, persistent memories by integrating temporally continuous and semantically related memory points, and during retrieval, it aligns with the temporal intent of the query to deliver contextually consistent and temporally grounded personalized responses. Evaluated on the LongMemEval and LoCoMo benchmarks, TSM significantly outperforms current methods, achieving up to a 12.2% improvement in accuracy, thereby demonstrating its effectiveness in accurately modeling long-term states and dynamic temporal patterns.

Technology Category

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

Application Category

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

Abstract

Memory enables Large Language Model (LLM) agents to perceive, store, and use information from past dialogues, which is essential for personalization. However, existing methods fail to properly model the temporal dimension of memory in two aspects: 1) Temporal inaccuracy: memories are organized by dialogue time rather than their actual occurrence time; 2) Temporal fragmentation: existing methods focus on point-wise memory, losing durative information that captures persistent states and evolving patterns. To address these limitations, we propose Temporal Semantic Memory (TSM), a memory framework that models semantic time for point-wise memory and supports the construction and utilization of durative memory. During memory construction, it first builds a semantic timeline rather than a dialogue one. Then, it consolidates temporally continuous and semantically related information into a durative memory. During memory utilization, it incorporates the query's temporal intent on the semantic timeline, enabling the retrieval of temporally appropriate durative memories and providing time-valid, duration-consistent context to support response generation. Experiments on LongMemEval and LoCoMo show that TSM consistently outperforms existing methods and achieves up to 12.2% absolute improvement in accuracy, demonstrating the effectiveness of the proposed method.

Problem

• Temporal Semantic Memory
• LLM agents
• personalization
• memory modeling
• temporal dimension

Innovation

• Temporal Semantic Memory
• durative memory
• semantic timeline
• personalized LLM agents
• temporal intent