Abstract Purpose Glutamine, as an essential part of enteral nutrition and parenteral nutrition agent, has been widely recognized to be a kind of important intestinal mucosa protectant in clinical practice and experimental research. However, the mechanisms of its protective effects are still not fully understand. Consequently, this study aimed to explore the potential mechanism of glutamine on ischemia-reperfusion (I/R) injury induced endoplasmic reticulum (ER) stress in intestine. Methods An experimental model of intestinal I/R in rats was established by 1 hour occlusion of the superior mesenteric artery followed by 3 hours of reperfusion. Morphologic changes of intestinal mucosa, apoptosis of epithelial cells, and expression of intestinal Grp78, Gadd153, Caspase-12, ATF4, PERK phosphorylation (P-PERK) and elF2αphosphorylation(P-elF2α) were determined. Results After I/R, the apoptotic index of intestinal mucosa epithelial cells observably increased with notable necrosis of intestinal mucosa, and the expressions of Grp78, Gadd153, Caspase-12, ATF4, P-PERK and P-elF2αall were increased. However, treatment with glutamine could significantly relieve intestinal I/R injury and apoptosis index. Moreover, glutamine could clearly up-regulate the expression of Grp78, restrain P-PERK and P-elF2α, and reduce ATF4, Gadd153 and Caspase-12 expressions. Conclusion Glutamine may be involved in alleviating ER stress induced intestinal mucosa cells apoptosis.

Il presente studio ha come obiettivo quello di misurare il valore di Voice Onset Time (VOT, l’intervallo di tempo tra l’esplosione del suono occlusivo e l’inizio delle vibrazioni periodiche del suono seguente) delle consonanti occlusive bilabiali e delle consonanti occlusive alveolari o dentali dell’italiano nella produzione orale dei parlanti cinesi. In questo esperimento, le produzioni dei parlanti sono state analizzate sotto forma di spettrogramma. I risultati dell’analisi hanno mostrato una forte diminuzione dei valori di VOT da parte di alcuni parlanti cinesi, spiegabile attraverso i successivi confronti tra occlusive sorde e sonore nelle due lingue. I sinofoni, invece di creare una nuova categoria fonetica, hanno cercato di rimpiazzare le occlusive sorde non aspirate presenti nella L1 con le occlusive sonore italiane perché questi due tipi di suoni, con alcune differenze, ricadono nella stessa categoria, definita short lag. Ciò dimostra che il VOT è un vincolo fonologico non facile da superare da parte di parlanti non nativi e rappresenta una proprietà fonetico-acustica che rimane salda nella produzione orale della L2.

Emotion-cause pair extraction (ECPE) aims to identify all emotion and cause clauses in documents, forming emotion-cause pairs (ECPs). Although existing methods have achieved some success, they face issues such as overlooking the impact of emotion experiencers, failing to leverage specific domain knowledge, and tending to spurious correlations. To address these issues, we transform the ECPE task into a multi-step reasoning problem and propose the Emotion-Experiencer-Event-Cause (EEEC) multi-step framework. EEEC combines symbolic prior reasoning with large language model (LLM) inference to improve precision and interpretability. Specifically, we design a lightweight symbolic filtering module that uses sentiment lexicons and interpretable rules to pre-select candidate emotion clauses, thereby reducing the reasoning space and guiding the LLM toward emotionally salient content. Furthemore, we introduce an experiencer identification task to understand the source of emotions and enhance the association between emotion and cause clauses. In addition, by combining both prior knowledge and induced reasoning, EEEC guides a large-scale language model (LLM) to perform the ECPE task efficiently. Experimental results demonstrate that EEEC achieves performance close to current state-of-the-art supervised fine-tuning methods.

Emotion-Cause Pair Extraction (ECPE) aims to identify clause-level pairs expressing emotions and their underlying causes. Existing methods rely on monolithic architectures that model causal relations through homogeneous cues, such as syntactic or sequential proximity, making them fragile to long-range or semantically com plex associations. We argue that emotion-cause reasoning is inherently heterogeneous, emerging from the confluence of diverse linguistic signals including syntactic, semantic role, coreference, temporal, and commonsense. To capture this complexity, we propose GraphEME, a multi-expert graph neural network that decomposes inference into specialized linguistic perspectives, each encoded as a subgraph and processed independently. A learnable gating mechanism dynamically fuses expert outputs, enhancing both robustness and interpretability. Experiments on Chinese ECPE benchmarks show that GraphEME achieves state-of-the-art performance, with ablation and case studies validating the necessity and effectiveness of multi-expert fusion.

Federated learning (FL) for skeleton-based action recognition remains underexplored, particularly under strong client heterogeneity where regular FedAvg tends to cause client drift and unstable convergence. We introduce Clustered Federated Spatio-Temporal Graph Attention Networks (CF-STGAT), a clustered FL framework that leverages attention-derived spatio-temporal statistics from local STGAT models to dynamically group clients and perform attention-weighted inter-cluster fusion that gently align cluster models. Concretely, the server periodically extracts multi-head parameter-based attention descriptors, normalizes and projects them via PCA, and applies K-means to form clusters; a global reference is then computed by attention–similarity weighting and used to regularize each cluster model with a lightweight fusion step. On NTU RGB+D 60/120(NTU 60/120), CF-STGAT consistently outperforms strong FL baselines with the STGAT backbone, yielding absolute top-1 gains of +0.84/+4.09 (NTU 60, X-Sub/X-Setup) and +7.98/+4.18 (NTU 120, X-Sub/X-Setup) over FedAvg, alongside smoother per-client trajectories and lower terminal test loss. Ablations indicate that attention-guided clustering and inter-cluster fusion are complementary: clustering reduces within-group variance whereas fusion limits cross-cluster divergence. The approach keeps local training unchanged and adds only server-side statistics and clustering.

Large-scale Pre-trained Language Models (PLMs) have become the backbones of text classification due to their exceptional performance. However, they treat input documents as independent and uniformly distributed, thereby disregarding potential relationships among the documents. This limitation could lead to some miscalculations and inaccuracies in text classification. To address this issue, some recent work explores the integration of Graph Neural Networks (GNNs) with PLMs, as GNNs can effectively model document relationships. Yet, combining graph-based methods with PLMs is challenging due to the structural incompatibility between graphs and sequences. To tackle this challenge, we propose a graph-enhanced text mutual learning framework that integrates graph-based models with PLMs to boost classification performance. Our approach separates graph-based methods and language models into two independent channels and allows them to approximate each other through mutual learning of probability distributions. This probability-distribution-guided approach simplifies the adaptation of graph-based models to PLMs and enables seamless end-to-end training of the entire architecture. Moreover, we introduce Asymmetrical Learning, a strategy that enhances the learning process, and incorporate Uncertainty Weighting loss to achieve smoother probability distribution learning. These enhancements significantly improve the performance of mutual learning. The practical value of our research lies in its potential applications in various industries, such as social network analysis, information retrieval, and recommendation systems, where understanding and leveraging document relationships are crucial. Importantly, our method can be easily combined with different PLMs and consistently achieves state-of-the-art results on multiple public datasets.

Learning motivation is a key variable in learning. Therefore, its assessment has consistently been a popular research topic. While traditional methods like self-report still dominate, methods integrating passive mobile sensing have emerged, using smartphones to collect behavioral data and assess learning motivation via statistical and machine learning techniques. Recent advances in large language models (LLMs) offer new perspectives for psychological measurements, yet their application in learning motivation assessment remains underexplored. To bridge this gap, we propose a novel approach that integrates LLMs with passive mobile sensing to assess and predict students' learning motivation. We constructed our dataset using mobile sensing data and self-report measures, then designed zero-shot and few-shot tasks embedded in LLMs to evaluate the performance. Ultimately, our findings indicate the feasibility and highlight the potential of leveraging LLMs to predict learning motivation levels based on mobile sensing data.

Despite the popularity of MOOC, only a small percentage of course participants complete the course. Learners' confusion is one of the factors that impacts the overall learning process and ultimately leads to course attrition. However, extensive exchange of reviews often creates chaos, resulting in 'confusion' posts that can easily be ignored. To this end, we create a labeled dataset for post-type classification and use the public Stanford MOOCPosts dataset to detect learning confusion. We propose a hierarchical multi-task learning framework that combines post-type and confusion degree classification using fine-tuned BERT models and virtual adversarial training. Our model performs with an accuracy of 89\% and an F1 score of 88\%. Additionally, we integrate interpretability techniques to further enhance model transparency. This framework equips instructors with tools to identify and address learning confusion effectively.

Emotion-cause pair extraction (ECPE) main focus is on extracting all potential emotion clauses and corresponding cause clauses from unannotated documents. Existing methods achieve promising results with the help of fine-tuning and prompt paradigms, but they present three downsides. First, most approaches cannot distinguish between the emotion-cause pairs that belong to different types of emotions, limiting the existing approaches' applicability. Second, existing prompt methods utilize a one-to-one mapping relation to achieve label words to category mapping, which brings considerable bias to the results. Third, existing methods achieve the cause extraction task supported by explicit semantic understanding or basic prompt templates, ignoring the implicit information contained in the cause clauses themselves. To solve these issues, we propose an Emotion knowledge-aware Prompt-tuning for Emotion-Cause Pair Extraction (EmoPrompt-ECPE) method, which integrate the knowledge of emotion categories in the ECPE task and mine the implicit knowledge of cause clauses. Specifically, we inject the latent knowledge of the cause clauses and the emotion types into the prompt template. Besides, we extend the emotion labels for many-to-one mapping of label words to categories with an external emotion word base. Furthermore, we utilize the cosine similarity filtering of the label word base to reduce the noise caused by knowledge introduction. Experiments on both Chinese and English benchmark datasets show that our approach can achieve state-of-the-art results. Our code and data can be found at: https://github.com/xy-xiaotudou/EmoPrompt-ECPE.

Understanding and identifying the nature of learner confusion is important for online learning platforms. In this study, we address this problem by analyzing forum posts from large-scale online courses. However, due to the large volume of comments and frequent interactions, confusion posts are often overlooked. Existing methods and models, while capable of detecting confusion, typically rely on linguistic features of posts and community factors (e.g. votes, views) but ignore personalized contexts, such as the specific causes and types of confusion. To address this problem, we create the first deep learning dataset focused on confusion types and develop a BERT-based network to model personalized features and identify confusion types. Considering the highly imbalanced distribution of different types of confusion, we further design a novel loss function that adaptively optimizes the training weights for each type. Our method’s effectiveness is confirmed through extensive experimentation.