Ying Xing

Hypertension is a leading contributor to cardiovascular morbidity worldwide. When blood pressure and routine clinical variables are available during clinic visits or health checks, guideline thresholds support binary classification but do not necessarily provide reliable calibrated probabilities for decision support. Machine learning studies in hypertension have concentrated on predicting future disease incidence rather than binary classification at the time of measurement, and calibration and validation design are reported inconsistently across this literature. This study examines whether machine learning, trained under a stratified protocol with held‐out calibration and test subsets, can improve probabilistic reliability in post‐measurement classification while remaining comparable to a clinical blood pressure rule. We analysed a public hypertension dataset of 4240 records with a positive‐class prevalence of 31.1% using a scikit‐learn pipeline that integrates preprocessing, recursive feature elimination, Optuna‐based hyperparameter tuning, and isotonic calibration on a held‐out calibration set. A tuned decision tree was compared against a clinical rule based on systolic blood pressure of at least 140 mmHg or diastolic blood pressure of at least 90 mmHg across 30 stratified random splits. Accuracy and recall were comparable to the clinical rule, with no statistically significant Wilcoxon differences, whereas Brier score and expected calibration error were significantly lower for the machine learning model, with mean values of 0.081 versus 0.090 and 0.027 versus 0.084. Interpretability analysis was consistent with blood‐pressure‐related predictors dominating model attributions. These findings suggest that machine learning classification after blood pressure measurement aligns with guideline‐based decisions while offering better calibrated probabilities for clinical decision support.

Background: Detecting software vulnerabilities and analyzing their impacts on interconnected software components are crucial to ensuring the security of software‐driven edge computing and Artificial Intelligence of Things (AIoT) systems, especially with the growing prevalence of software supply chain attacks. Knowledge graphs (KGs) have been used to model complex semantic dependencies between vulnerabilities and software components. However, existing KGs often contain noisy or irrelevant relations, which significantly hinder their utility in downstream tasks. Objective: This study aims to improve vulnerability detection and impact analysis by denoising vulnerability KGs and generating task‐oriented vulnerability representations. Methods: We propose a diffusion‐based graph learning framework for denoising vulnerability KGs. The framework first employs a diffusion process to reconstruct relational structures within the KG, guided by a graph alignment mechanism that integrates collaborative signals from the software component–vulnerability bipartite graph. This alignment encourages the denoised KG to preserve relations most relevant to downstream tasks. Subsequently, a heterogeneous information network aggregates multi‐relational information from the denoised KG to produce knowledge‐enhanced embeddings for vulnerabilities. Results: Through this two‐stage learning paradigm, the framework effectively captures both semantic and relational dependencies among heterogeneous entities and generates task‐oriented representations for vulnerability detection and impact analysis. Experimental results on a widely used benchmark demonstrate that the framework can effectively identify latent relationships between vulnerabilities and software components and predict high‐risk vulnerabilities. Conclusion: The proposed framework improves the utility of vulnerability KGs by suppressing noisy relations and preserving task‐relevant information, providing practical insights for proactive vulnerability mitigation and patch recommendation.

Cross-Project Defect Prediction (CPDP) utilizes the existing labeled data in the source project to assist with the prediction of unlabeled projects in the target dataset, which effectively improves the prediction performance and has become a research hotspot in software engineering. At present, CPDP can be categorized into homogeneous CPDP and heterogeneous CPDP (HDP), in which HDP doesn’t require that the source project and the target project have the same feature space, thus, it is more widely used in the actual CPDP. Most of current HDP methods map the original features to the latent feature space and reduce the inter-project variation by transferring domain-independent features, but the transferring process ignores the use of domain-related features, which affects the prediction performance of the model. Moreover, the mapped latent features are not conducive to the model’s interpretability. Based on these, this article proposes a Heterogeneous Defect Prediction method based on Feature Disentanglement (FD-HDP). We disentangle the features using domain-related and domain-independent feature extractors, respectively, to improve the interpretability of the model by maximizing the domain adversarial loss during training and guiding the feature extractors to produce accurate domain-related and domain-independent features. The weighted sum of the prediction results from domain-related and domain-independent predictors is used as the final prediction result of the project during the prediction process, which realizes the combination of domain-independent and domain-related features and effectively improves the prediction performance. In this article, we conducted experiments using four publicly available defect datasets to construct heterogeneous scenarios. The results demonstrate that the FD-HDP model shows significant advantages over state-of-the-art methods in six metrics.

Dense audio-visual event localization (DAVE) aims to identify event categories and locate the temporal boundaries in untrimmed videos. For such challenging task settings, most studies only employ audio-visual event semantic constraints on the final outputs, lacking progressive cross-modal semantic bridging in intermediate layers. This causes semantic gaps that hinder alignment between representations in audio and visual features, making it difficult to distinguish between event-related and irrelevant background content. Moreover, they rarely consider the correlations between events, which limits the model to infer co-occurring events among complex scenarios. In this paper, we incorporate multi-stage semantic guidance and multi-event relationship modeling, which respectively enable progressive semantic understanding of audio-visual events and adaptive extraction of event dependencies, thereby better focusing on event-related information. Specifically, our event-aware semantic guided network (ESG-Net) includes a early semantic interaction (ESI) module and a mixture of dependency experts (MoDE) module. ESI applys multi-stage semantic guidance to explicitly constrain the model in learning semantic information through multi-stage feature fusion and several classification loss functions, ensuring multi-stage understanding of event-related content. MoDE promotes the extraction of multi-event dependencies through multiple serial mixture of experts with adaptive weight allocation. Extensive experiments demonstrate that our method significantly surpasses the state-of-the-art methods, while greatly reducing parameters and computational load. Our code will be released on https://github.com/uchiha99999/ESG-Net.

Bug fixing holds significant importance in software development and maintenance. Recent research has made substantial strides in exploring the potential of large language models (LLMs) for automatically resolving software bugs. However, a noticeable gap in existing approaches lies in the oversight of collaborative facets intrinsic to bug resolution, treating the process as a single-stage endeavor. Moreover, most approaches solely take the buggy code snippet as input for LLMs during the patch generation stage. To mitigate the aforementioned limitations, we introduce a novel stage-wise framework named PATCH. Specifically, we first augment the buggy code snippet with corresponding dependence context and intent information to better guide LLMs in generating the correct candidate patches. Additionally, by taking inspiration from bug management practices, we decompose the bug-fixing task into four distinct stages: bug reporting, bug diagnosis, patch generation, and patch verification. These stages are performed interactively by LLMs, aiming to simulate the collaborative behavior of programmers during the resolution of software bugs. By harnessing these collective contributions, PATCH effectively enhances the bug-fixing capability of LLMs. We implement PATCH by employing the powerful dialogue-based LLM ChatGPT. Our evaluation on the widely used bug-fixing benchmark BFP demonstrates that PATCH has achieved better performance than state-of-the-art LLMs.

Fine-grained image categorization aims to categorize subclasses by processing detailed features, which is still a critical problem to be solved in computer version due to the small differences between subclasses. The traditional methods are usually to find features by manual annotation, using specific sliding Windows, using different thresholds and other methods. These methods are not only costly, but also ineffective. In computer version, by calculating attention scores between parts of the picture multiple times and weighting them, the transformer greatly improves the accuracy of categorization. In this paper, we propose a feature weight units. Specifically, transformer is used as the backbone to capture image feature(these features are called patches in transformer), and then all patches are weighted by our feature weight unit. The computal result of feature fusion unit represents the importance of the patch should to be forced on. To verify the effectiveness of our method, we conducted experiments on the CUB-200-2011 and stanford-dog datasets.