Variational autoencoder (VAE)-based frameworks possess a natural advantage in modeling the shared and private information inherent in multimodal data. However, current models focus on improving the quality of shared representations from the reconstruction perspective, lacking explicit mechanisms to model their underlying semantic structure. In this paper, we propose the multimodal Gaussian mixture variational autoencoder with consistency regularizations, which introduces a Gaussian mixture prior over the shared latent space to enhance its semantic structure and encourage the formation of cluster-aware latent representations. To address the cross-modal inconsistency problem under missing modality conditions, we propose a cluster-guided regularization strategy that enforces the cross-modal consistency using the pseudo-category labels from unsupervised clustering. Additionally, we design a self-supervised contrastive regularization strategy to align semantically similar representations across modalities. Extensive experiments on MNIST-SVHN and MNIST-CDCB datasets demonstrate that our method significantly outperforms prior state-of-the-art models in generation, classification, and retrieval tasks.

Personalized mobile network service is a hotpot issue now. In order to provide the real-time and accurate personalized mobile network service, researchers introduce the trust into the mobile user need model. However the existing research rarely considers the context information and the structure of mobile social network when calculating the trust. Hence, in this paper, we propose a trust calculating algorithm which can improve the accuracy of the trust in mobile social network. Firstly, we analyze mobile users' behaviors and incorporate the context into the trust calculating model. Secondly, the trust is introduced into the division of mobile community, and an improved method of mobile community division is proposed. Thirdly, according to the obtained mobile community structure, we propose a power calculating method and merge the trust and the obtained power. Finally, the similarity of contextual mobile user preferences is introduced into the trust. The experimental results show that our method can get more accurate community division and trust.