Cross-Modal Attention Network with Dual Graph Learning in Multimodal Recommendation
ACM Transactions on Multimedia Computing, Communications, and Applications · 31.03.2026 · 22 (3) · 1-23
DOI: 10.1145/3788289 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Multimedia Computing, Communications, and Applications, online erschienen am 27.02.2026. Verfasst von Ji Dai und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 37 Literaturangaben. Gefördert von: Beijing Natural Science Foundation, National Natural Science Foundation of China, Open Fund of the Key Laboratory for Civil Aviation Collaborative Air Traffic Management Technology and Applications.
Abstract (Originalsprache)
Multimedia recommendation systems leverage user–item interactions and multimodal information to capture user preferences, enabling more accurate and personalized recommendations. Despite notable advancements, existing approaches still face two critical limitations: first, shallow modality fusion often relies on simple concatenation, failing to exploit rich synergic intra- and inter-modal relationships; second, asymmetric feature treatment—where users are only characterized by interaction IDs while items benefit from rich multimodal content—hinders the learning of a shared semantic space. To address these issues, we propose a C ross-Modal R ecursive A ttention N etwork with Dual Graph E mbedding (CRANE) . To tackle shallow fusion, we design a core Recursive Cross-Modal Attention (RCA) mechanism that iteratively refines modality features based on cross-correlations in a joint latent space, effectively capturing high-order intra- and inter-modal dependencies. For symmetric multimodal learning, we explicitly construct users’ multimodal profiles by aggregating features of their interacted items. Furthermore, CRANE integrates a symmetric dual graph framework—comprising a heterogeneous user–item interaction graph and a homogeneous item–item semantic graph—unified by a self-supervised contrastive learning objective to fuse behavioral and semantic signals. Despite these complex modeling capabilities, CRANE maintains high computational efficiency. Theoretical and empirical analyses confirm its scalability and high practical efficiency, achieving faster convergence on small datasets and superior performance ceilings on large-scale ones. Comprehensive experiments on four public real-world datasets validate an average 5% improvement in key metrics over state-of-the-art baselines. Our code is publicly available at https://github.com/MKC-Lab/CRANE .
Förderung
- Beijing Natural Science Foundation
- National Natural Science Foundation of China
- Open Fund of the Key Laboratory for Civil Aviation Collaborative Air Traffic Management Technology and Applications
- SMP-Z Large Model Fund
Zitieren
APA
Dai, J., Fang, Q., Hu, J., Cai, D., Yang, Y., & Zhao, C. (2026). Cross-Modal Attention Network with Dual Graph Learning in Multimodal Recommendation. ACM Transactions on Multimedia Computing, Communications, and Applications, 22(3), 1-23. https://doi.org/10.1145/3788289
Vancouver
Dai J, Fang Q, Hu J, Cai D, Yang Y, Zhao C. Cross-Modal Attention Network with Dual Graph Learning in Multimodal Recommendation. ACM Transactions on Multimedia Computing, Communications, and Applications. 2026;22(3):1-23. doi:10.1145/3788289
BibTeX
@article{Dai2026,
author = {Dai, Ji and Fang, Quan and Hu, Jun and Cai, Desheng and Yang, Yang and Zhao, Can},
title = {Cross-Modal Attention Network with Dual Graph Learning in Multimodal Recommendation},
journal = {ACM Transactions on Multimedia Computing, Communications, and Applications},
year = {2026},
volume = {22},
number = {3},
pages = {1--23},
doi = {10.1145/3788289}
}
EUVIMED