Structurally Refined Graph Transformer for Multimodal Recommendation

TL;DR

This paper introduces SRGFormer, a novel multimodal recommendation model that leverages hypergraph structures and self-supervised learning to better capture user preferences and improve recommendation accuracy.

Contribution

The paper proposes a structurally refined transformer model that integrates multimodal data into hypergraphs and employs self-supervised tasks to enhance recommendation performance.

Findings

SRGFormer outperforms previous models by 4.47% on the Sports dataset.

Embedding multimodal info into hypergraphs improves local structure learning.

Self-supervised tasks enhance the integration of user-item collaborative signals.

Abstract

Multimodal recommendation systems utilize various types of information, including images and text, to enhance the effectiveness of recommendations. The key challenge is predicting user purchasing behavior from the available data. Current recommendation models prioritize extracting multimodal information while neglecting the distinction between redundant and valuable data. They also rely heavily on a single semantic framework (e.g., local or global semantics), resulting in an incomplete or biased representation of user preferences, particularly those less expressed in prior interactions. Furthermore, these approaches fail to capture the complex interactions between users and items, limiting the model's ability to meet diverse users. To address these challenges, we present SRGFormer, a structurally optimized multimodal recommendation model. By modifying the transformer for better integration into our model, we capture the overall behavior patterns of users. Then, we enhance structural information by embedding multimodal information into a hypergraph structure to aid in learning the local structures between users and items. Meanwhile, applying self-supervised tasks to user-item collaborative signals enhances the integration of multimodal information, thereby revealing the representational features inherent to the data's modality. Extensive experiments on three public datasets reveal that SRGFormer surpasses previous benchmark models, achieving an average performance improvement of 4.47 percent on the Sports dataset. The code is publicly available online.