Self-Supervised Learning (SSL) methods harness the concept of semantic invariance by utilizing data augmentation strategies to produce similar representations for different deformations of the same input. Essentially, the model captures the shared information among multiple augmented views of samples, while disregarding the non-shared information that may be beneficial for downstream tasks. To address this issue, we introduce a module called CompMod with Meta Comprehensive Regularization (MCR), embedded into existing self-supervised frameworks, to make the learned representations more comprehensive. Specifically, we update our proposed model through a bi-level optimization mechanism, enabling it to capture comprehensive features. Additionally, guided by the constrained extraction of features using maximum entropy coding, the self-supervised learning model learns more comprehensive features on top of learning consistent features. In addition, we provide theoretical support for our proposed method from information theory and causal counterfactual perspective. Experimental results show that our method achieves significant improvement in classification, object detection and semantic segmentation tasks on multiple benchmark datasets.

Higher education plays a critical role in cultivating talent, preserving culture, and promoting social progress. However, current challenges, such as inefficient information dissemination and low problem-solving efficiency among students, highlight the need for intelligent question-answering systems. These systems, leveraging artificial intelligence and natural language processing technologies, enable rapid and accurate responses to student queries, thereby providing intelligent support for higher education management. This study introduces the ATT-MR-WL model, a generative AI system integrating Mask R-CNN and Word2Vec+LSTM to enhance intelligent question-answering functionality. The model, customized to handle both text and visual data, is evaluated using the established VQA v2.0 dataset and a specially developed EM dataset reflecting university management scenarios. The ATT-MR-WL model demonstrates a 3% accuracy improvement over traditional methods and enhances its ability to handle multimodal queries. This research provides important insights for enhancing the efficiency and quality of higher education management and advancing the process of educational informatization.

Human motion prediction (HMP) has emerged as a popular research topic due to its diverse applications. Traditional methods rely on hand-crafted features and machine learning techniques, which often struggle to model the complex dynamics of human motion. Recent deep learning-based methods have achieved success by learning spatio-temporal representations of motion, but these models often overlook the reliability of motion data. Additionally, the temporal and spatial dependencies of skeleton nodes are distinct. The temporal relationship captures motion information over time, while the spatial relationship describes body structure and the relationships between different nodes. In this paper, we propose a novel spatio-temporal branching network using incremental information for HMP, which decouples the learning of temporal-domain and spatial-domain features, extracts more motion information, and achieves complementary cross-domain knowledge learning through knowledge distillation. Our approach effectively reduces noise interference and provides more expressive information for characterizing motion by separately extracting temporal and spatial features. We evaluate our approach on standard HMP benchmarks and outperform state-of-the-art methods in terms of prediction accuracy. Code is available at https://github.com/JasonWang959/STPMP.