EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Xiaoxiong, Zhou, Xin, Zeng, Zhiwei, Wang, Yongjie, Niyato, Dusit, Shen, Zhiqi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909748773781504
author Zhang, Xiaoxiong
Zhou, Xin
Zeng, Zhiwei
Wang, Yongjie
Niyato, Dusit
Shen, Zhiqi
author_facet Zhang, Xiaoxiong
Zhou, Xin
Zeng, Zhiwei
Wang, Yongjie
Niyato, Dusit
Shen, Zhiqi
contents MultiModal Recommendation (MMR) systems have emerged as a promising solution for improving recommendation quality by leveraging rich item-side modality information, prompting a surge of diverse methods. Despite these advances, existing methods still face two critical limitations. First, they use raw modality features to construct item-item links for enriching the behavior graph, while giving limited attention to balancing collaborative and modality-aware semantics or mitigating modality noise in the process. Second, they use a uniform alignment weight across all entities and also maintain a fixed alignment strength throughout training, limiting the effectiveness of modality-behavior alignment. To address these challenges, we propose EGRA. First, instead of relying on raw modality features, it alleviates sparsity by incorporating into the behavior graph an item-item graph built from representations generated by a pretrained MMR model. This enables the graph to capture both collaborative patterns and modality aware similarities with enhanced robustness against modality noise. Moreover, it introduces a novel bi-level dynamic alignment weighting mechanism to improve modality-behavior representation alignment, which dynamically assigns alignment strength across entities according to their alignment degree, while gradually increasing the overall alignment intensity throughout training. Extensive experiments on five datasets show that EGRA significantly outperforms recent methods, confirming its effectiveness.
format Preprint
id arxiv_https___arxiv_org_abs_2508_16170
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation
Zhang, Xiaoxiong
Zhou, Xin
Zeng, Zhiwei
Wang, Yongjie
Niyato, Dusit
Shen, Zhiqi
Information Retrieval
Artificial Intelligence
MultiModal Recommendation (MMR) systems have emerged as a promising solution for improving recommendation quality by leveraging rich item-side modality information, prompting a surge of diverse methods. Despite these advances, existing methods still face two critical limitations. First, they use raw modality features to construct item-item links for enriching the behavior graph, while giving limited attention to balancing collaborative and modality-aware semantics or mitigating modality noise in the process. Second, they use a uniform alignment weight across all entities and also maintain a fixed alignment strength throughout training, limiting the effectiveness of modality-behavior alignment. To address these challenges, we propose EGRA. First, instead of relying on raw modality features, it alleviates sparsity by incorporating into the behavior graph an item-item graph built from representations generated by a pretrained MMR model. This enables the graph to capture both collaborative patterns and modality aware similarities with enhanced robustness against modality noise. Moreover, it introduces a novel bi-level dynamic alignment weighting mechanism to improve modality-behavior representation alignment, which dynamically assigns alignment strength across entities according to their alignment degree, while gradually increasing the overall alignment intensity throughout training. Extensive experiments on five datasets show that EGRA significantly outperforms recent methods, confirming its effectiveness.
title EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2508.16170