Generative artificial intelligence (GAI) has demonstrated remarkable potential in both educational practice and research, particularly in areas such as personalized learning, adaptive assessment, innovative teaching methods, and crosscultural communication. However, it faces several significant challenges, including the comprehension of complex domain knowledge, technological accessibility, and the delineation of AI's role in education. Addressing these challenges necessitates collaborative efforts from educators and researchers. This paper summarizes the state-of-the-art large language models developed by various technology companies, exploring their diverse applications and unique contributions to primary, higher, and vocational education. Furthermore, it reviews recent research from the past three years, focusing on the challenges and solutions associated with GAI in educational practice and research. The aim of the review is to provide novel insights for enhancing human-computer interaction in educational settings through the utilization of GAI. Statistical analysis reveals that the current application of large language models in the education sector is predominantly centered on the ChatGPT series. A key focus for future research lies in effectively integrating a broader range of large language models into educational tasks, with particular emphasis on the interaction between multimodal large language models and educational scenarios

Most session-based recommender systems (SBRSs) focus on extracting information from the observed items in the current session of a user to predict a next item, ignoring the causes outside the session (called outer-session causes, OSCs) that influence the user’s selection of items. However, these causes widely exist in the real world, and few studies have investigated their role in SBRSs. In this work, we analyze the causalities and correlations of the OSCs in SBRSs from the perspective of causal inference. We find that the OSCs are essentially the confounders in SBRSs, which leads to spurious correlations in the data used to train SBRS models. To address this problem, we propose a novel SBRS framework named COCO-SBRS (COunterfactual COllaborative Session-Based Recommender Systems) to learn the causality between OSCs and user-item interactions in SBRSs. COCO-SBRS first adopts a self-supervised approach to pre-train a recommendation model by designing pseudo-labels of causes for each user’s selection of the item in data to guide the training process. Next, COCO-SBRS adopts counterfactual inference to recommend items based on the outputs of the pre-trained recommendation model considering the causalities to alleviate the data sparsity problem. As a result, COCO-SBRS can learn the causalities in data, preventing the model from learning spurious correlations. The experimental results of our extensive experiments conducted on three real-world datasets demonstrate the superiority of our proposed framework over ten representative SBRSs.

Complex networks enable to represent and characterize the interactions between entities in various complex systems which widely exist in the real world and usually generate vast amounts of data about all the elements, their behaviors and interactions over time. The studies concentrating on new network analysis approaches and methodologies are vital because of the diversity and ubiquity of complex networks. The stochastic block model (SBM), based on Bayesian theory, is a statistical network model. SBMs are essential tools for analyzing complex networks since SBMs have the advantages of interpretability, expressiveness, flexibility and generalization. Thus, designing diverse SBMs and their learning algorithms for various networks has become an intensively researched topic in network analysis and data mining. In this article, we review, in a comprehensive and in-depth manner, SBMs for different types of networks (i.e., model extensions), existing methods (including parameter estimation and model selection) for learning optimal SBMs for given networks and SBMs combined with deep learning. Finally, we provide an outlook on the future research directions of SBMs.

Multimodal recommender systems (MRSs) aim to integrate information from multiple modalities, for better capturing users’ preferences. However, existing MRSs usually face the challenge of data sparsity, especially for the limited user-item interactions. It is not uncommon that a user interacts with only a few items out of millions. More importantly, the interaction information is easy to be overwhelmed, leading to poor recommendation performance. To tackle these significant gaps for further improving the performance of MRSs, we propose a Modality-behavior Fusion Model Integrated with Graph Contrastive Learning for Multimodal Recommendation (FGCM). First, through the introduction of random noise into graph contrastive learning, FGCM adeptly mitigates data sparsity. Then, a modality-behavior fuser is proposed to effectively fuse interaction data and multimodal side information data by taking cooperative signals and multimodal signals as the guidance. Extensive experiments on three datasets demonstrate that our proposed approach outperforms several state-of-the-art methods.