Information Fusion最新文献

In practical multi-view applications, some data in each view are missing. Although recent incomplete multi-view clustering (IMC) approaches have achieved encouraging performance, two challenges remain. They utilize the tensor nuclear norm to explore the high-order correlations among view-specific similarity graphs. Moreover, they only infer the missing views but do not recover the consensus cluster structure across complete views. To address these issues, we propose a new method called graph Refinement and consistency Self-Supervision for Tensorized Incomplete Multi-view Clustering (RS-TIMC). Specifically, RS-TIMC introduces graph decomposition to remove the diverse similarities from the view-specific graphs and utilizes the tensor Schatten-p norm to model the consistent parts. Additionally, by extracting features from the original observable data and inferring the missing instances, RS-TIMC enables the cluster structure of each complete view to be adjusted. Finally, RS-TIMC utilizes consistent similarity graphs to recover the shared local geometric structure across all complete views. Experimental evaluations on several datasets indicate that our method outperforms the start-of-the-art approaches.

In Natural Language Processing (NLP), state-of-the-art machine learning models heavily depend on vast amounts of training data. Often, this data is sourced from third parties, such as crowdsourcing platforms, to enable swift and efficient annotation collection for supervised learning. Yet, such an approach is susceptible to poisoning attacks where malicious agents deliberately insert harmful data to skew the resulting model behavior. Current countermeasures to these attacks either come at a significant cost, lack full efficacy, or are simply non-applicable. This study introduces and evaluates the potential of personalized model architectures as a defense against these threats. By comparing two top-performing personalized model architectures, User-ID and HuBi-Medium, against a standard non-personalized baseline across two NLP tasks and various simulated attack scenarios, we found that the personalized model architectures significantly outperformed the baseline. The robustness advantage increased with the rise in malicious annotations. Notably, the User-ID model excelled in safeguarding predictions for legitimate users from the influence of malicious annotations. Our findings emphasize the benefit of adopting personalized model architectures to bolster NLP system defenses against poisoning attacks.

Multimodal Emotion Recognition (MER) involves integrating information of various modalities, including audio, visual, text and physiological signals, to comprehensively grasp human sentiments, which has emerged as a vibrant area within human–computer interaction. Researchers have developed many methods for this task, but many of these methods rely on labeled supervised learning and struggle to address the issue of missing some modalities of data. To address these issues, we propose a Multiplex Graph Aggregation and Feature Refinement framework for unsupervised incomplete MER, comprising four modules: Completion, Aggregation, Refinement, and Embedding. Specifically, we first capture the correlation information between samples using the graph structures, which aids in the completion of missing data and the multiplex aggregation of multimodal data. Then, we perform refinement operations on the aggregated features as well as alignment and enhancement operations on the embedding features to obtain the fused feature representations, which are consistent, highly separable and conducive to emotion recognition. Experimental results on multimodal emotion recognition datasets demonstrate that our method achieves state-of-the-art performance among unsupervised methods, validating its effectiveness.