M²VAE: Multi-Modal Multi-View Variational Autoencoder for Cold-start Item Recommendation
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (17) · 14811-14819
DOI: 10.1609/aaai.v40i17.38501 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Chuan He und 5 weiteren Autor:innen.
Abstract (Originalsprache)
Cold-start item recommendation is a significant challenge in recommendation systems, particularly when new items are introduced without any historical interaction data. While existing methods leverage multi-modal content to alleviate the cold-start issue, they often neglect the inherent multi-view structure of modalities, namely the distinction between shared and modality-specific features. In this paper, we propose Multi-Modal Multi-View Variational AutoEncoder (M²VAE), a generative model that addresses the challenges of modeling common and unique views in attribute and multi-modal features, as well as user preferences over single-typed item features. Specifically, we generate type-specific latent variables for item IDs, categorical attributes, and image features, and use Product-of-Experts (PoE) to derive a common representation. A disentangled contrastive loss decouples the common view from unique views while preserving feature informativeness. To model user inclinations, we employ a user-aware hierarchical Mixture-of-Experts (MoE) to adaptively fuse representations. We further incorporate co-occurrence signals via contrastive learning, eliminating the need for pretraining. Extensive experiments on real-world datasets validate the effectiveness of our approach.
Zitieren
APA
He, C., Liu, Y., Li, Q., Hong, C., Zhong, W., & Yao, X. (2026). M²VAE: Multi-Modal Multi-View Variational Autoencoder for Cold-start Item Recommendation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(17), 14811-14819. https://doi.org/10.1609/aaai.v40i17.38501
Vancouver
He C, Liu Y, Li Q, Hong C, Zhong W, Yao X. M²VAE: Multi-Modal Multi-View Variational Autoencoder for Cold-start Item Recommendation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(17):14811-14819. doi:10.1609/aaai.v40i17.38501
BibTeX
@article{He2026,
author = {He, Chuan and Liu, Yongchao and Li, Qiang and Hong, Chuntao and Zhong, Wenliang and Yao, Xin-Wei},
title = {M²VAE: Multi-Modal Multi-View Variational Autoencoder for Cold-start Item Recommendation},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {17},
pages = {14811--14819},
doi = {10.1609/aaai.v40i17.38501}
}
EUVIMED