Turning Semantics into Topology: LLM-Driven Attribute Augmentation for Collaborative Filtering

TL;DR

This paper introduces TAGCF, a novel framework that leverages large language models to convert semantic knowledge into topological structures, enhancing collaborative filtering by modeling interaction intents and causal relations within an enriched graph.

Contribution

The paper proposes a new method that transforms semantic signals into topological connectivity using LLMs and introduces an adaptive graph convolution to improve recommendation accuracy.

Findings

Consistent performance improvements across multiple datasets.

Effective handling of cold-start scenarios.

Robustness validated through extensive experiments.

Abstract

Large Language Models (LLMs) have shown great potential for enhancing recommender systems through their extensive world knowledge and reasoning capabilities. However, effectively translating these semantic signals into traditional collaborative embeddings remains an open challenge. Existing approaches typically fall into two extremes: direct inference methods are computationally prohibitive for large-scale retrieval, while embedding-based methods primarily focus on unilateral feature augmentation rather than holistic collaborative signal enhancement. To bridge this gap, we propose Topology-Augmented Graph Collaborative Filtering (TAGCF), a novel framework that transforms semantic knowledge into topological connectivity. Unlike existing approaches that depend on textual features or direct interaction synthesis, TAGCF employs LLMs to infer interaction intents and underlying causal relationships from user-item pairs, representing these insights as intermediate attribute nodes within an enriched User-Attribute-Item (U-A-I) graph. Furthermore, to effectively model the heterogeneous relations in this augmented structure, we propose Adaptive Relation-weighted Graph Convolution (ARGC), which employs relation-specific prediction networks to dynamically estimate the importance of each relation type. Extensive experiments across multiple benchmark datasets and CF backbones demonstrate consistent improvements, with comprehensive evaluations including cold-start scenarios validating the effectiveness and robustness of our framework. All code will be made publicly available. For anonymous review, our code is available at the following anonymous link: https://anonymous.4open.science/r/AGCF-2441353190/.