Depth estimation is a critical component of many computer vision applications, enabling accurate spatial awareness from visual inputs. It is particularly vital in autonomous driving and robotics, where precise depth information supports essential functions such as navigation and obstacle detection. Traditional methods, including sensor-based and monocular vision techniques, often face limitations such as high costs or reduced accuracy. In contrast, binocular depth estimation, which infers depth by analyzing disparities between stereo image pairs, offers a compelling balance of cost-effectiveness and precision. However, many existing models suffer from high computational demands and long inference times, limiting their suitability for real-time deployment. To address these challenges, we propose a streamlined convolutional neural network optimized for binocular depth estimation. Our model significantly reduces architectural complexity while maintaining strong performance. Additionally, we introduce a novel loss function that incorporates adaptive weighting and consistency constraints to enhance accuracy and stability during training. We evaluate our method on the KITTI 2015 benchmark and demonstrate that it achieves competitive accuracy while significantly reducing runtime compared to existing approaches. These improvements make our method a practical and efficient solution for real-time depth estimation in resource-constrained environments.

Depth estimation is a core task in computer vision with applications in autonomous driving, robotics, augmented reality, and 3D reconstruction. In recent years, deep learning has significantly advanced depth prediction, enabling both single-image and multi-image approaches to achieve impressive accuracy across diverse environments. This paper presents a comprehensive review of current deep learning-based depth estimation methods, categorized by input type: single-image and multi-image. We examine key architectural developments, representative models, and benchmark performance on widely used datasets, particularly KITTI and NYU Depth-v2. Despite notable progress, current methods face critical limitations in real-world deployment, including high computational cost and limited robustness to environmental variation. These issues are especially concerning for safety-critical applications where real-time inference and stability are essential. We conclude by highlighting open challenges and encouraging future research toward efficient, robust, and deployable depth estimation systems.

In recent years, the adoption of cloud services has been expanding at an unprecedented rate. As more organizations migrate or deploy their businesses to the cloud, a multitude of related cybersecurity incidents, such as data breaches, are on the rise. Several inherent attributes of cloud environments, including data sharing, remote access, dynamic scalability, and scalability, pose significant challenges for the protection of cloud security. Even more concerning is the growing threat of Advanced Persistent Threats (APTs), which have become increasingly sophisticated and stealthy. As a more complex form of multi-stage attacks (MSAs), APTs follow a multi-step process that spreads malicious actions across different stages and blends them with legitimate operations, making intrusion detection particularly challenging. In this paper, we conduct an empirical study on the multi-stage characteristics of APT attacks specifically within cloud environments. Drawing from real-world attack scenarios, we analyze the behavioral patterns of APT attacks, evaluate the existing countermeasures, and identify ongoing challenges in defending against them. Our findings expose the limitations of conventional intrusion detection approaches and highlight the critical need for fine-grained and behavior-aware security mechanisms specifically designed for cloud environments. This study provides a deeper understanding of how APTs adapt their tactics to exploit cloud-specific vulnerabilities and offers insights for improving threat detection and response in modern cloud infrastructures.

Brain tumors, characterized by the uncontrolled growth of cells within the brain, pose a formidable challenge in medical diagnostics due to their complex nature and the critical need for precise intervention. Magnetic Resonance Imaging (MRI) plays a pivotal role in the early detection and classification of brain tumors, offering detailed insights into tumor types without invasive procedures. However, the manual interpretation of MRI scans is labor-intensive, subject to human error, and heavily dependent on the expertise of radiologists. These challenges underscore the urgent need for advanced computational tools that enhance diagnostic accuracy and efficiency. This paper introduces XcepKNN, a novel architecture that integrates a K-Nearest Neighbor (KNN) classifier within the Xception deep learning framework, specifically designed to improve the classification of brain tumor MRI images. The hybrid model leverages the depthwise separable convolutions of Xception to extract detailed features from MRI data, while the embedded KNN classifier utilizes these features to accurately identify and classify various brain tumor types. The fusion of these techniques facilitates a more nuanced analysis of MRI images, enhancing the model’s ability to distinguish between tumor categories with high precision. Our extensive validation on a dataset of 7,023 MRI images demonstrates that XcepKNN significantly outperforms traditional models in terms of accuracy, precision, recall, and F1 scores. By providing an open-source implementation, this study contributes to the field of medical image analysis, offering a reliable tool for researchers and clinicians alike to improve the diagnostic processes for brain tumors.

Employee turnover in the restaurant industry poses significant operational and financial challenges, primarily due to the high costs associated with hiring and training new staff, as well as the negative impact on team cohesion and customer service. This paper employs a comprehensive datadriven approach using machine learning (ML) and deep learning (DL) techniques to analyze employee turnover, aiming to identify and understand the key predictors and underlying patterns. Utilizing a dataset of 778 responses from restaurant employees, the study explores various factors, including job satisfaction, income levels, and perceptions of management. Through rigorous data preprocessing and the application of advanced analytical models such as Decision Trees, Random Forests, and Deep Neural Networks, the study enhances the predictive accuracy and reliability of turnover predictions. The results not only highlight the effectiveness of specific models but also shed light on significant turnover predictors. This research contributes to the development of targeted strategies for reducing turnover rates, ultimately aiding in improving both operational efficiency and employee retention in the restaurant industry.

In this study, we present a novel attention-based encoder-decoder model for monocular depth estimation, show-casing exceptional robustness and accuracy across standard and noise-injected variants of the KITTI dataset. By inte-grating Convolutional Block Attention Module (CBAM) and Squeeze-and-Excitation (SE) blocks, our approach significantly outperforms existing state-of-the-art methods, especially in en-vironments affected by real-world noise such as changes in brightness, saturation, and RGB channels. The model's superior performance, validated through rigorous testing, marks a signif-icant step forward in the field, offering promising applications in autonomous driving, augmented reality, and beyond. Our work demonstrates the potential of attention mechanisms in enhancing depth estimation models to reliably interpret complex scenes, paving the way for advancements in depth-dependent technologies operating in dynamic and challenging conditions.

Convolutional neural networks (CNNs) have gained increasing popularity and versatility in recent decades, finding applications in diverse domains. These remarkable achievements are greatly attributed to the support of extensive datasets with precise labels. However, annotating image datasets is intricate and complex, particularly in the case of multi-label datasets. Hence, the concept of partial-label setting has been proposed to reduce annotation costs, and numerous corresponding solutions have been introduced. The evaluation methods for these existing solutions have been primarily based on accuracy. That is, their performance is assessed by their predictive accuracy on the test set. However, we insist that such an evaluation is insufficient and one-sided. On one hand, since the quality of the test set has not been evaluated, the assessment results are unreliable. On the other hand, the partial-label problem may also be raised by undergoing adversarial attacks. Therefore, incorporating robustness into the evaluation system is crucial. For this purpose, we first propose two attack models to generate multiple partial-label datasets with varying degrees of label missing rates. Subsequently, we introduce a lightweight partial-label solution using pseudo-labeling techniques and a designed loss function. Then, we employ D-Score to analyze both the proposed and existing methods to determine whether they can enhance robustness while improving accuracy. Extensive experimental results demonstrate that while certain methods may improve accuracy, the enhancement in robustness is not significant, and in some cases, it even diminishes.

Multi-label classification is a widely encountered problem in daily life, where an instance can be associated with multiple classes. In theory, this is a supervised learning method that requires a large amount of labeling. However, annotating data is time-consuming and may be infeasible for huge labeling spaces. In addition, label imbalance can limit the performance of multi-label classifiers, especially when some labels are missing. Therefore, it is meaningful to study how to train neural networks using partial labels. In this work, we address the issue of label imbalance and investigate how to train classifiers using partial labels in large labeling spaces. First, we introduce the pseudo-labeling technique, which allows commonly adopted networks to be applied in partially labeled settings without the need for additional complex structures. Then, we propose a novel loss function that leverages statistical information from existing datasets to effectively alleviate the label imbalance problem. In addition, we design a dynamic training scheme to reduce the dimension of the labeling space and further mitigate the imbalance. Finally, we conduct extensive experiments on some publicly available multi-label datasets such as COCO, NUS-WIDE, CUB, and Open Images to demonstrate the effectiveness of the proposed approach. The results show that our approach outperforms several state-of-the-art methods, and surprisingly, in some partial labeling settings, our approach even exceeds the methods trained with full labels.

As open source software is widely used, the vulnerabilities contained therein are also rapidly propagated to a large number of innocent applications. Even worse, many vulnerabilities in open-source projects are secretly fixed, which leads to affected software being unaware and thus exposed to risks. For the purpose of protecting deployed software, designing an effective patch classification system becomes more of a need than an option. To this end, some researchers take advantage of the recent advancements in natural language processing to learn both commit messages and code changes. However, they often incur high false positive rates. Not only that, existing works cannot yet answer how much the textual description (such as commit messages) alone can influence the final triage. In this paper, we propose a Transformer based patch classifier, which does not use any code changes as inputs. Surprisingly, the extensive experiment shows the proposed approach can significantly outperform other state-of-the-art work with a high precision of 93.0% and low false positive rate. Therefore, our research further confirms the critical importance of well-crafted commit messages for the later software maintenance. Finally, our case study also identifies 48 silent security patches, which can benefit those affected software.

BACKGROUND: The prognosis and survival of lung adenocarcinoma (LUAD) patients are still not promising despite recent breakthroughs in treatment. Endoplasmic reticulum stress (ERS) is a self-protective mechanism resulting from an imbalance in quality control of unfolded proteins when cells are stressed, which plays an active role in lung cancer development, but the relationship between ERS and the pathological characteristics and clinical prognosis of LUAD patients remains unclear. METHODS: LASSO and Cox regression were applied based on sequencing information to construct the model, which was validated to be robust. The risk scores of the patients were calculated using the formula provided by the model, and the patients were divided into high and low-risk groups according to the median cut-off of risk scores. Cox regression analysis identifies independent prognostic factors for these patients, and enrichment analysis of prognosis-related genes was also performed. The relationship between risk scores and tumor mutation burden (TMB), cancer stem cell index, and drug sensitivity was explored. RESULTS: We constructed a 13-gene prognostic model for LUAD patients. Patients in the high-risk group had worse overall survival, lower immune score and ESTIMATE score, higher TMB, higher cancer stem cell index, and higher sensitivity to conventional chemotherapeutic agents. In addition, we constructed a nomogram that predicts 5-year survival in LUAD patients, which helps clinicians to foresee the prognosis from a new perspective. CONCLUSIONS: Our results highlight the association of ERS with LUAD and the potential use of ERS in guiding treatment.

In this article, test diversity has been suggested to be a valid way to improve test suite effectiveness. Extended finite state machine (EFSM) is a widely used formal model, but little attention is paid on the test suite generation with more diversity. EFSM test suite generation involves test paths generation and test data generation. Considering the discrepancy between test paths has a more crucial impact on the diversity of test suite, compared with the difference between test data, this article, therefore, mainly concerns the test paths generation with more diversity for EFSM models. Hence, the factors that influence the discrepancy between test paths are investigated. Then based on these factors, an integrated distance metric is designed to evaluate the dissimilarity between test paths, and a diversity measurement for EFSM test suite is presented. Furthermore, a diversity-oriented test suite generation (DOTSG) method is proposed where a dissimilarity-based fitness function and diversity-oriented update strategy are adopted in traditional coverage-oriented EFSM test suite generation (COTSG) by genetic algorithm. The experimental results show that, compared to COTSG, our DOTSG can not only generate more diverse test suite to satisfy a certain coverage criteria, improving the fault detection capability of the test suite, but also decrease the evolution time cost and the size of test suite generated.

Online music recommender systems are becoming increasingly prevalent because of the popularity of digital music and music recommendation generally caters to users by discovering songs that match their preferences. However, these systems have to face a challenge: how to recommend new songs in a situation where prior knowledge is scarce. Some researches take auxiliary information into consideration in new recommendation approaches to deal with this problem. Nevertheless, they rarely pay attention to complex relationships among different feature spaces when they map those information to a latent space. To this end, this paper proposes an approach that uses non-linear transformation to integrate different auxiliary information into the songs latent representations. Unlike other studies which directly map auxiliary information to the feature space, the proposed music recommendation model (NeuTrans) maps different information features to low-dimensional vector representation by non-linear neural networks. Specifically, the NeuTrans separately employs matrix factorization and attribute network embedding to extract auxiliary information (historical interaction, network structure and attributes of songs). The feature space of different information is obtained by nonlinearly mapping the feature space of the songs. Experimental analysis on two real-world datasets shows that our framework outperforms the state-of-the- art approaches on Top-N music recommendation.

With the growing popularity of online social platforms, it has been universally recognized that incorporating social relations into recommender systems can usually alleviate the problem of data sparsity. However, social recommender systems based on explicit relations are not as successful as expected due to the noise and the social cold issue of explicit social links. The intuition of utilizing explicit relations is that users share similar preferences if they are friends in the social network. In fact, quite a lot of users who are distant from each other in the social network also have similar tastes. The user item network and the user social network can provide useful information that can complement each other, so that exploring the implicit friends using the heterogeneous network they formed would be more helpful. In this paper, we propose an approach IFSR to discover implicit friends over the heterogeneous network to improve the performance of social recommendation. To find out reliable implicit ties, we first model the system as a heterogeneous network upon which both the preferences and social information are coupled. Over the HIN, similarities between each pair of users can be quantified through network embedding based representation learning. To reduce the computational cost while preserving the information embedded in the original networks and uncover the latent information hiding in the HIN, several meaningful meta-paths over the HIN are designed to guide the process of random walks. Finally, the Top-K implicit friends are incorporated into a social bayesian ranking model to enhance the performance of item ranking. Experimental results on three datasets demonstrate IFSR outperforms the state-of-the-art methods and illustrate why the implicit friends are advantageous for social recommendation.

The recommendation systems have been widely employed due to the effectiveness on mitigating the information overload issue. At present, the recommendation systems have made great progress, but they are under the threat of shilling attack because of their open nature. Shilling attack is the way by which the attackers can manipulate the recommendation results and cause great harm to recommendation systems. Existing shilling attack detection models are mainly based on statistical measures to extract features like the rating deviation, which are generally susceptible to attack strategies. Once the attacker changes attack strategy, the detection model which is based on the statistical method may fail. Some researchers have identified that implicit features hidden in user-user interactions and user-item interactions can be utilized to solve the problem. Their research aims to learn potential relationship between users to update features. However, the research ignores the significance of learning features by employing label information. To solve this problem, in this paper, we propose a novel detection model, named BayesDetector, which takes not only the user-user and user-item interactions but also the label information into consideration in the process of learning user implicit features. Furthermore, to take full advantage of user labels, the Bayesian model is added to the feature learning. Experiments on two datasets, Amazon and Movielens, show that BayesDetector significantly outperforms the state-of-the-art methods.