Traditional feature pyramid structures have limitations in cross-level semantic information fusion, leading to inconsistent detection results across different scales. This paper proposes a Mutual Information-based Feature Pyramid Enhancement mechanism (MI-FPE) to address the lack of consensus between feature pyramid levels in real-time object detection. The aim is to improve the performance of Real-Time Detection Transformer (RT-DETR) models in multi-scale object detection. MI-FPE uses a scale alignment module to unify feature map sizes. It employs a mutual information maximization strategy to extract common semantic information across levels. An adaptive fusion mechanism then flexibly injects this common information into each original feature layer. Experimental results on the MS-COCO dataset show significant improvements in detection performance, especially for small objects. With ResNet-50 as the backbone network,$A P_{\text{val }}$reaches 53.2%, a 0.1 percentage point increase over the baseline RT-DETR. The small object detection metric$A P_{S}^{\text{val }}$improves by 1.8 percentage points to 36.6%.

Hyper-relational knowledge graphs (HKGs) extend the traditional triplet-based knowledge graph by adding qualifiers to the relationships, making HKGs particularly useful for tasks that require more profound understanding and inference from relationships between entities. However, existing hyper-relational knowledge representation learning methods (HKRL) focus on direct neighbourhood information of entities only by neglecting the relational similarity of the main triple in hyper-relational facts and the attribute details in the qualifiers. In addition, few works extract common and private information across multiple views to minimize noise and interference. This paper proposes a multi-hypergraph disentanglement method for HKRL to address the above issues. Specifically, we first construct four hypergraphs to mine and utilise the inherent structure information of HKGs, and then propose to extract common representations among hypergraphs and private representations within individual hypergraphs to mine the semantic information and the task-relevant information, respectively. Experiment results on four real datasets demonstrate the effectiveness of the proposed method compared to SOTA methods in link prediction tasks on HKGs.

In deep contrastive graph clustering, many methods tend to adopt a singular enhancement strategy, focusing either on structural or attribute augmentation. This limited approach constrains the model's ability to integrate multidimensional information, resulting in an imbalance in information utilization. Furthermore, the randomness involved in the selection of negative samples may lead to blurred distinctions between positive and negative samples. Therefore, we introduce a graph attention-based dual enhancement multiview clustering (GA-DE-MVC). The GA-DE-MVC first encodes structure and attributes through attention mechanisms and fully connected layers to achieve dual enhancement of structure and attributes. Then, it selects the farthest sample as the negative sample based on the Euclidean distance between cluster centers, to alleviate the randomness in negative sample selection, thereby enhancing the distinctiveness between positive and negative samples.The experimental results on six datasets surpass existing algorithms, verifying the effectiveness of the GA-DE-MVC algorithm.