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

2026-02-24

AI Summary

This work addresses the challenge of effectively integrating the semantic knowledge of large language models (LLMs) with the topological structure inherent in collaborative filtering for recommendation systems. To this end, we propose the Topology-Augmented Graph Collaborative Filtering (TAGCF) framework, which uniquely models user–item interaction intents and causal relationships inferred by LLMs as explicit attribute nodes within a heterogeneous user–attribute–item (U–A–I) graph. We further introduce an Adaptive Relation-weighted Graph Convolution (ARGC) mechanism to organically fuse semantic knowledge with collaborative signals. Extensive experiments demonstrate that TAGCF consistently achieves significant performance gains across multiple benchmark datasets and various backbone models, exhibiting particularly strong robustness and effectiveness in cold-start scenarios.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

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

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/.

Problem

• Large Language Models
• Collaborative Filtering
• Semantic-to-Topological Transformation
• Recommender Systems
• Graph-based Recommendation

Innovation

• LLM-driven topology
• attribute-augmented graph
• collaborative filtering
• adaptive relation weighting
• causal intent inference