Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Federated Personal Knowledge Graph Completion with Lightweight Large Language Models for Personalized Recommendations

Fernando Spadea, Oshani Seneviratne

Abstract

Personalized recommendation increasingly relies on private user data, motivating approaches that can adapt to individuals without centralizing their information. We present Federated Targeted Recommendations with Evolving Knowledge graphs and Language Models (FedTREK-LM), a framework that unifies lightweight large language models (LLMs), evolving personal knowledge graphs (PKGs), federated learning (FL), and Kahneman-Tversky Optimization to enable scalable, decentralized personalization. By prompting LLMs with structured PKGs, FedTREK-LM performs context-aware reasoning for personalized recommendation tasks such as movie and recipe suggestions. Across three lightweight Qwen3 models (0.6B, 1.7B, 4B), FedTREK-LM consistently and substantially outperforms state-of-the-art KG completion and federated recommendation baselines (HAKE, KBGAT, and FedKGRec), achieving more than a 4x improvement in F1-score on the movie and food benchmarks. Our results further show that real user data is critical for effective personalization, as synthetic data degrades performance by up to 46%. Overall, FedTREK-LM offers a practical paradigm for adaptive, LLM-powered personalization that generalizes across decentralized, evolving user PKGs.

Federated Personal Knowledge Graph Completion with Lightweight Large Language Models for Personalized Recommendations

Abstract

Personalized recommendation increasingly relies on private user data, motivating approaches that can adapt to individuals without centralizing their information. We present Federated Targeted Recommendations with Evolving Knowledge graphs and Language Models (FedTREK-LM), a framework that unifies lightweight large language models (LLMs), evolving personal knowledge graphs (PKGs), federated learning (FL), and Kahneman-Tversky Optimization to enable scalable, decentralized personalization. By prompting LLMs with structured PKGs, FedTREK-LM performs context-aware reasoning for personalized recommendation tasks such as movie and recipe suggestions. Across three lightweight Qwen3 models (0.6B, 1.7B, 4B), FedTREK-LM consistently and substantially outperforms state-of-the-art KG completion and federated recommendation baselines (HAKE, KBGAT, and FedKGRec), achieving more than a 4x improvement in F1-score on the movie and food benchmarks. Our results further show that real user data is critical for effective personalization, as synthetic data degrades performance by up to 46%. Overall, FedTREK-LM offers a practical paradigm for adaptive, LLM-powered personalization that generalizes across decentralized, evolving user PKGs.
Paper Structure (35 sections, 8 figures, 3 tables)

This paper contains 35 sections, 8 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Contributions:
  3. Related Work
  4. Recommendation using KG Completion in Centralized Settings.
  5. LLMs for Personalized Recommendations.
  6. Federated Model Adaptation.
  7. System Architecture
  8. Structured Prompting with PKGs
  9. Knowledge Representation
  10. Federated Training Process
  11. KTO Application
  12. Experimental Setup and Training Protocol
  13. Dataset Construction
  14. Real Movie Data Construction.
  15. Synthetic Data Generation.
  16. ...and 20 more sections

Figures (8)

  • Figure 1: Overall performance results of our TREK-LM models against the baselines (KBGAT, HAKE, and FedKGRec).
  • Figure 2: Overview of the FedTREK-LM architecture.
  • Figure 3: Detailed view of the FedTREK-LM local operations: (a) shows the feedback-driven fine-tuning using KTO, and (b) illustrates how the LLM reasons over the PKG to derive recommendations.
  • Figure 4: Illustration of KTO data construction in the form of prompt–completion–label triplets from user-LLM interactions.
  • Figure 5: TREK-LM Performance by Setting (Qwen3-4B)
  • ...and 3 more figures