Beyond Interactions: Node-Level Graph Generation for Knowledge-Free Augmentation in Recommender Systems
This work addresses the need for efficient and applicable recommender systems without external resources, offering a novel augmentation method that is incremental in advancing knowledge-free approaches.
The paper tackles the problem of data dependency and computational overhead in recommender systems by proposing NodeDiffRec, a knowledge-free augmentation framework that generates node-level graphs to enhance recommendations, achieving up to 98.6% improvement in Recall@5 and 84.0% in NDCG@5 over baselines.
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.