PersonaAgent with GraphRAG: Community-Aware Knowledge Graphs for Personalized LLM

2025-11-21

AI Summary

This paper addresses the limitations of existing AI agent personalization frameworks in modeling user “personas” and capturing contextual awareness. To this end, we propose a knowledge graph–enhanced, community-aware personalized language modeling framework. Methodologically: (1) We construct a community-aware user knowledge graph by integrating LLM-generated graph indices with community-level behavioral summaries, enabling joint modeling of individual personas and collective knowledge; (2) We design a dynamic context-aware prompt generation mechanism that synergistically leverages Graph RAG, graph neural networks, and community detection algorithms to jointly reason over historical interactions and global interaction patterns. On the LaMP benchmark, our approach achieves +11.1% F1 in news classification, +56.1% F1 in movie tagging, and −10.4% MAE in product rating prediction—substantially outperforming state-of-the-art personalized RAG and fine-tuning baselines. Our core contributions are the first incorporation of community structure into user knowledge graph construction and the unification of persona-driven dynamic graph retrieval and generation.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Natural Language Processing: (Large) Language Models
• Humans and AI: Human-Aware Planning and Behavior Prediction

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
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

We propose a novel framework for persona-based language model system, motivated by the need for personalized AI agents that adapt to individual user preferences. In our approach, the agent embodies the user's "persona" (e.g. user profile or taste) and is powered by a large language model (LLM). To enable the agent to leverage rich contextual information, we introduce a Knowledge-Graph-enhanced Retrieval-Augmented Generation (Graph RAG) mechanism that constructs an LLM-derived graph index of relevant documents and summarizes communities of related information. Our framework generates personalized prompts by combining: (1) a summary of the user's historical behaviors and preferences extracted from the knowledge graph, and (2) relevant global interaction patterns identified through graph-based community detection. This dynamic prompt engineering approach allows the agent to maintain consistent persona-aligned behaviors while benefiting from collective knowledge. On the LaMP benchmark, our method improves news categorization F1 by 11.1%, movie tagging F1 by 56.1%, and reduces product rating MAE by 10.4% over prior methods. Our code is available at https://anonymous.4open.science/r/PersonaAgentwGraphRAG-DE6F

Problem

• Creating personalized AI agents adapting to individual user preferences
• Enhancing LLM responses with knowledge graphs and community detection
• Generating dynamic prompts combining personal history and global patterns

Innovation

• Graph RAG constructs knowledge graphs from documents
• Community detection identifies global interaction patterns
• Dynamic prompts combine personal and collective knowledge