With the rapid development of social media, sentiment analysis from multimodal posts has garnered significant attention in recent years. However, the substantial size of these models impedes their deployment on resour...
详细信息
Crowdsourcing testing can reduce testing costs and improve testing efficiency. However, crowdsourcing testing generates many test cases, from which testers need to select task-related test cases for execution. Further...
详细信息
This paper proposes a semi-supervised learning framework for cardiac image segmentation based on a bicycle Variational Auto-Encoder (biVAE) architecture by embedding a Prior Transformer and a conditional generative ne...
详细信息
Unit testing aims to validate the correctness of software system units and has become an essential practice in software development and maintenance. However, it is incredibly time-consuming and labor-intensive for tes...
详细信息
Unit testing aims to validate the correctness of software system units and has become an essential practice in software development and maintenance. However, it is incredibly time-consuming and labor-intensive for testing experts to write unit test cases manually, including test inputs (i.e., prefixes) and test oracles (i.e., assertions). Very recently, some techniques have been proposed to apply Large Language Models (LLMs) to generate unit assertions and have proven the potential in reducing manual testing efforts. However, there has been no systematic comparison of the effectiveness of these LLMs, and their pros and cons remain unexplored. To bridge this gap, we perform the first extensive study on applying various LLMs to automated assertion generation. The experimental results on two independent datasets show that studied LLMs outperform six state-of-the-art techniques with a prediction accuracy of 51.82%-58.71% and 38.72%-48.19%. The improvements achieve 29.60% and 12.47% on average. Besides, as a representative LLM, CodeT5 consistently outperforms all studied LLMs and all baselines on both datasets, with an average improvement of 13.85% and 26.64%, respectively. We also explore the performance of generated assertions in detecting real-world bugs, and find LLMs are able to detect 32 bugs from Defects4J on average, with an improvement of 52.38% against the most recent approach EDITAS. Inspired by the findings, we construct a simplistic retrieval-and-repair- enhanced LLM-based approach by transforming the assertion generation problem into a program repair task for retrieved similar assertions. Surprisingly, such a simplistic approach can further improve the prediction accuracy of LLMs by 9.40% on average, leading to new records on both datasets. Besides, we provide additional discussions from different aspects (e.g., the impact of assertion types and test lengths) to illustrate the capacity and limitations of LLM-based approaches. Finally, we further pinpoint va
Accurate localization is critical for lunar rovers exploring lunar terrain features. Traditionally, lunar rover localization relies on sensor data from odometers, inertial measurement units and stereo cameras. However...
详细信息
Human skeleton point data has better environmental adaptability and motion expression ability than RGB video data. Therefore, the action recognition algorithm based on skeletal point data has received more and more at...
详细信息
To improve the accuracy of steel surface defect detection, this study proposes an improved multi-directional optimization model based on the YOLOv10n algorithm. First, we introduce innovations to the convolution (C2F)...
详细信息
The increased use of cyberspace and technological advancement are fundamentally changing the cyber threat landscape. Cyberattacks are becoming more sophisticated, frequent, and destructive. Internationally, there is a...
详细信息
ISBN:
(纸本)9781914587405
The increased use of cyberspace and technological advancement are fundamentally changing the cyber threat landscape. Cyberattacks are becoming more sophisticated, frequent, and destructive. Internationally, there is a growing acceptance that Cyber Counterintelligence (CCI) is essential to counter cyber-attacks optimally. Therefore, in addition to government intelligence and security agencies, more companies are incorporating a CCI approach as a critical element of their posture for engaging cyber threats. However, the successful adoption of a CCI approach depends on the availability of skilled CCI professionals equipped with the requisite competences. The creation of such CCI professionals, in turn, requires a framework for developing the necessary CCI competences. At least in as far as reviewed academic literature is concerned, there is no existing postulation on a framework to develop the CCI competences, specifically for developing countries. Given the complexity and multi-disciplinary nature of the emerging CCI field, such a framework needs to provide two distinctive skillsets linked to CCI’s two distinct areas of expertise, namely cyber (security) and counterintelligence. The paper presents a high-level Cyber Counterintelligence Competence Framework (CCIC Framework) that outlines dimensions of CCI, functional areas, job roles and requisite competences (knowledge, skills, and abilities), and tasks for each CCI job role. The CCI framework also outlines five levels of proficiency expected for each job role. The identification of competences and levels of proficiency are integral to the successful implementation of the framework and workforce development. The CCIC Framework is intended to be used as a tool to retain, assess, and monitor knowledge, skills, and abilities for CCI workforce development. In addition, the CCIC Framework can be used to assist in providing the basis for individual performance management, education, training, and development pathway, as wel
Accurately predicting oil well production volume is of great significance in oilfield production. To overcome the shortcomings in the current study of oil well production prediction, we propose a hybrid model (GRU-KAN...
详细信息
Accurately predicting oil well production volume is of great significance in oilfield production. To overcome the shortcomings in the current study of oil well production prediction, we propose a hybrid model (GRU-KAN) with the gated recurrent unit (GRU) and Kolmogorov-Arnold network (KAN). The GRU-KAN model utilizes GRU to extract temporal features and KAN to capture complex nonlinear relationships. First, the MissForest algorithm is employed to handle anomalous data, improving data quality. The Pearson correlation coefficient is used to select the most significant features. These selected features are used as input to the GRU-KAN model to establish the oil well production prediction model. Then, the Particle Swarm Optimization (PSO) algorithm is used to enhance the predictive performance. Finally, the model is evaluated on the test set. The validity of the model was verified on two oil wells and the results on well F14 show that the proposed GRU-KAN model achieves a Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Coefficient of Determination (R2) values of 11.90, 9.18, 6.0% and 0.95, respectively. Compared to popular single and hybrid models, the GRU-KAN model achieves higher production-prediction accuracy and higher computational efficiency. The model can be applied to the formulation of oilfield-development plans, which is of great theoretical and practical significance to the advancement of oilfield technology levels.
A new interdigital spoof surface plasmon polaritons (SSPPs) unit is proposed to design dual-and triple-wideband filters in this letter. The dispersion properties of the proposed interdigital SSPPs are analyzed and ind...
详细信息
A new interdigital spoof surface plasmon polaritons (SSPPs) unit is proposed to design dual-and triple-wideband filters in this letter. The dispersion properties of the proposed interdigital SSPPs are analyzed and indicating that the asymptotic frequency of the fundamental mode (Mode 0) of the transition SSPPs unit is higher than that of the first-order mode (Mode 1) of the transmission SSPPs unit, which provides the feasibility in the design of dual passbands using Modes 0 and 1 of the SSPPs simultaneously. Also, by adjusting the stopband between Modes 0 and 1 of the transition interdigital SSPPs unit reasonably, three passbands can further be constructed. For verification, dual-and triple-wideband filters are accordingly well designed, fabricated, and measured. Good agreement between the simulations and measurements validities the proposed structure and design method.
暂无评论