PersonalAI: Towards digital twins in the graph form

2025-06-20

AI Summary

Large language models (LLMs) struggle to model long-term, personalized user history and generate temporally aware responses. Method: This paper proposes a hypergraph-based digital twin modeling framework for users. It introduces a composite knowledge graph integrating standard edges and two types of hyperedges, enabling LLMs to autonomously construct and dynamically update user representations. Built upon an extended AriGraph hypergraph architecture, the framework incorporates retrieval-augmented generation (RAG), temporal dialogue modeling, and robust reasoning over contradictory statements to ensure persistent memory of personalized knowledge and consistent inference. Contribution/Results: Evaluated on TriviaQA, HotpotQA, and an enhanced version of DiaASQ, the method demonstrates stable question-answering accuracy under temporal constraints and contradictory inputs, confirming strong robustness in knowledge graph construction and sustained personalization capability.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The challenge of personalizing language models, specifically the ability to account for a user's history during interactions, is of significant interest. Despite recent advancements in large language models (LLMs) and Retrieval Augmented Generation that have enhanced the factual base of LLMs, the task of retaining extensive personal information and using it to generate personalized responses remains pertinent. To address this, we propose utilizing external memory in the form of knowledge graphs, which are constructed and updated by the LLM itself. We have expanded upon ideas of AriGraph architecture and for the first time introduced a combined graph featuring both standard edges and two types of hyperedges. Experiments conducted on the TriviaQA, HotpotQA and DiaASQ benchmarks indicates that this approach aids in making the process of graph construction and knowledge extraction unified and robust. Furthermore, we augmented the DiaASQ benchmark by incorporating parameters such as time into dialogues and introducing contradictory statements made by the same speaker at different times. Despite these modifications, the performance of the question-answering system remained robust, demonstrating the proposed architecture's ability to maintain and utilize temporal dependencies.

Problem

• Personalizing language models using user history
• Retaining personal information for personalized responses
• Utilizing knowledge graphs for robust memory management

Innovation

• Knowledge graphs as external memory
• Combined graph with hyperedges
• Robust temporal dependency handling