Multivariate time series anomaly detection (MTSAD) is a critical task for real-time risk control and fault diagnosis in industrial monitoring, aerospace, and financial domains. Unsupervised MTSAD confronts three core challenges: label scarcity in practical scenarios, diverse anomaly patterns that demand adaptive modeling, and weak feature discriminability between normal and anomalous samples. To address these challenges, we propose a Prototype Memory and Contrastive Learning Based Unsupervised Anomaly Detection for Multivariate Time Series method named PC-UAD. PC-UAD comprises three core modules with hierarchical functionalities: (1) A Temporal PatchEmbedder, which adopts learnable positional encoding for dynamic temporal representation and incorporates channel projection to model adaptive cross-sensor dependencies in multivariate data; (2) A Prototype Memory Encoder, which embeds a prototype attention mechanism to explicitly memorize typical normal patterns, forming a “normal pattern dictionary” that enhances the model’s perception of normal behavioral boundaries; (3) A ContrastFusion module, which leverages contrastive learning to amplify feature distribution discrepancies between normal and anomalous data, strengthening the model’s ability to distinguish subtle anomalies. Experiments on five public multivariate time series datasets demonstrate that our method achieves superior detection accuracy compared to eight state-of-the-art approaches, with the average F1-score and ROC-AUC both ranking first.

Visual Question Answering (VQA) with external knowledge requires external knowledge and visual content to answer questions about images. The defect of existing VQA solutions is that they need to identify task-related information in the obtained pictures, questions, and knowledge graphs. It is necessary to properly fuse and embed the information between different modes identified, to reduce the noise and difficulty in cross-modality reasoning of VQA models. However, this process of rationally integrating information between different modes and joint reasoning to find relevant evidence to correctly predict the answer to the question still deserves further study. This paper proposes a bimodal Graph Neural Network model combining pre-trained Language Models and Knowledge Graphs (BIGNN-LM-KG). Researchers built the concepts graph by the images and questions concepts separately. In constructing the concept graph, we used the combined reasoning advantages of LM+KG. Specifically, use KG to jointly infer the images and question entity concepts to build a concept graph. Use LM to calculate the correlation score to screen the nodes and paths of the concept graph. Then, we form a visual graph from the visual and spatial features of the filtered image entities. We use the improved GNN to learn the representation of the two graphs and to predict the most likely answer by fusing the information of two different modality graphs using a modality fusion GNN. On the common dataset of VQA, the model we proposed obtains good experiment results. It also verifies the validity of each component in the model and the interpretability of the model.