Beyond Interactions: Node-Level Graph Generation for Knowledge-Free Augmentation in Recommender Systems

2025-07-28

AI Summary

Existing recommendation enhancement methods either rely on external knowledge—such as knowledge graphs or large language models—suffering from data dependency and high computational overhead, or operate without external knowledge but lack fine-grained augmentation primitives, failing to bridge semantic and structural gaps. To address this, we propose NodeDiffRec: the first knowledge-free, node-level graph generative framework for recommendation enhancement. It synthesizes pseudo-items via a diffusion process and models denoising preferences directly on the user–item interaction graph, enabling fine-grained semantic enrichment and improved structural connectivity. Its core innovation lies in pioneering the application of diffusion models to unsupervised graph generative recommendation enhancement. Extensive experiments on multiple benchmark datasets demonstrate substantial improvements over state-of-the-art methods: Recall@5 increases by up to 98.6%, and NDCG@5 by up to 84.0%.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Graph-based Machine Learning
• Computer Vision: Diffusion Models for Vision

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: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Recent advances in recommender systems rely on external resources such as knowledge graphs or large language models to enhance recommendations, which limit applicability in real-world settings due to data dependency and computational overhead. Although knowledge-free models are able to bolster recommendations by direct edge operations as well, the absence of augmentation primitives drives them to fall short in bridging semantic and structural gaps as high-quality paradigm substitutes. Unlike existing diffusion-based works that remodel user-item interactions, this work proposes NodeDiffRec, a pioneering knowledge-free augmentation framework that enables fine-grained node-level graph generation for recommendations and expands the scope of restricted augmentation primitives via diffusion. By synthesizing pseudo-items and corresponding interactions that align with the underlying distribution for injection, and further refining user preferences through a denoising preference modeling process, NodeDiffRec dramatically enhances both semantic diversity and structural connectivity without external knowledge. Extensive experiments across diverse datasets and recommendation algorithms demonstrate the superiority of NodeDiffRec, achieving State-of-the-Art (SOTA) performance, with maximum average performance improvement 98.6% in Recall@5 and 84.0% in NDCG@5 over selected baselines.

Problem

• Enhance recommendations without external knowledge dependency
• Bridge semantic and structural gaps in knowledge-free models
• Generate node-level graph for improved diversity and connectivity

Innovation

• Node-level graph generation via diffusion
• Synthesizing pseudo-items for distribution alignment
• Denoising preference modeling for enhanced connectivity