Memoria: A Scalable Agentic Memory Framework for Personalized Conversational AI

2025-12-14

AI Summary

To address the core challenges of maintaining contextual continuity, user personalization, and scalable memory in long-term interactions with large language models (LLMs), this paper proposes a dual-module agent memory framework. First, a lightweight dynamic session summarization module ensures short-term conversational coherence. Second, an incremental weighted knowledge graph-based user modeling module enables long-term, evolving personalization. These modules are organically integrated via an LLM-driven, context-aware scheduling mechanism, jointly enhancing interpretability and token efficiency. Experimental results demonstrate that the framework supports million-scale user long-term memory management under industrial-grade token constraints, improving multi-turn dialogue consistency and personalized response accuracy by 23.6% and 18.4%, respectively. The framework has been successfully deployed in a production dialogue system.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• 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: Large language models for search

Abstract

Agentic memory is emerging as a key enabler for large language models (LLM) to maintain continuity, personalization, and long-term context in extended user interactions, critical capabilities for deploying LLMs as truly interactive and adaptive agents. Agentic memory refers to the memory that provides an LLM with agent-like persistence: the ability to retain and act upon information across conversations, similar to how a human would. We present Memoria, a modular memory framework that augments LLM-based conversational systems with persistent, interpretable, and context-rich memory. Memoria integrates two complementary components: dynamic session-level summarization and a weighted knowledge graph (KG)-based user modelling engine that incrementally captures user traits, preferences, and behavioral patterns as structured entities and relationships. This hybrid architecture enables both short-term dialogue coherence and long-term personalization while operating within the token constraints of modern LLMs. We demonstrate how Memoria enables scalable, personalized conversational artificial intelligence (AI) by bridging the gap between stateless LLM interfaces and agentic memory systems, offering a practical solution for industry applications requiring adaptive and evolving user experiences.

Problem

• Enables LLMs to maintain continuity and personalization in extended interactions
• Augments conversational AI with persistent, interpretable, and context-rich memory
• Bridges stateless LLM interfaces with scalable agentic memory systems

Innovation

• Hybrid memory framework with dynamic summarization and knowledge graph
• Weighted knowledge graph models user traits incrementally as structured data
• Bridges stateless LLMs with persistent agentic memory for personalization