Joint Behavior-guided and Modality-coherence Conditional Graph Diffusion Denoising for Multi Modal Recommendation

TL;DR

This paper introduces JBM-Diff, a diffusion model for multimodal recommendation that denoises features and feedback to improve ranking accuracy, addressing redundant information and feedback bias.

Contribution

It proposes a joint diffusion approach conditioned on collaborative features and modal consistency, enhancing multimodal feature alignment and data credibility detection.

Findings

Effective removal of irrelevant multimodal information.

Improved recommendation accuracy demonstrated on three datasets.

Enhanced data augmentation through sample pair credibility assessment.

Abstract

In recent years, multimodal recommendation has received significant attention and achieved remarkable success in GCN-based recommendation methods. However, there are two key challenges here: (1) There is a significant amount of redundant information in multimodal features that is unrelated to user preferences. Directly injecting multimodal features into the interaction graph can affect the collaborative feature learning between users and items. (2) There are false negative and false positive behaviors caused by system errors such as accidental clicks and non-exposure. This feedback bias can affect the ranking accuracy of training sample pairs, thereby reducing the recommendation accuracy of the model. To address these challenges, this work proposes a Joint Behavior-guided and Modal-consistent Conditional Graph Diffusion Model (JBM-Diff) for joint denoising of multimodal features and user feedback. We design a diffusion model conditioned on collaborative features for each modal feature to remove preference-irrelevant information, and enhance the alignment between collaborative features and modal semantic information through multi-view message propagation and feature fusion. Finally, we detect the partial order consistency of sample pairs from a behavioral perspective based on learned modal preferences, set the credibility for sample pairs, and achieve data augmentation. Extensive experiments on three public datasets demonstrate the effectiveness of this work. Codes are available at https://github.com/pxcstart/JBMDiff.