DRARec: diffusion-based representation augmentation for multimodal recommendation

Lingfeng Liu

doi:10.1007/s00530-026-02622-1

Return

DRARec: diffusion-based representation augmentation for multimodal recommendation

2026-09-06

PRE

DOI：10.1007/s00530-026-02622-1

Original

Original request for help

Save

Abstract

En 中文

Multimodal recommendation systems integrate multimodal information into user and item feature representations, and leverage multimodal information to alleviate the data sparsity issue. However, existing research faces two key challenges. First, most methods based on graph convolutional network (GCN) directly utilize multimodal features, which may introduce noisy information unrelated to user preferences. Second, existing diffusion-based methods primarily focus on interaction modeling or graph structure augmentation, which inevitably adds new noisy edges (e.g., introducing false user–item interactions). Although these methods have achieved remarkable success, the aforementioned noise can be amplified in the message propagation of GCN, affecting the final multimodal representation. To address these challenges, we propose a novel Diffusion-based Representation Augmentation for Multimodal Recommendation (DRARec) method. Specifically, to mitigate the negative impact of noisy information across different modalities, we introduce a behavior-aware injection mechanism. It first injects behavior modality information into all content modality embeddings, aligning them to a preference-related semantic space, and then employs an attention module to enhance feature discriminability. Additionally, to avoid the negative impact of noisy edges, we propose a diffusion-based modality representation augmentation strategy. It directly refines the semantic features of each modality to enhance the robustness of features. Extensive experiments on three public datasets demonstrate the superiority and effectiveness of DRARec.

Keywords:

Multimodal recommendation systems

Representation augmentation

Diffusion models