Abstract Drug–target interaction (DTI) prediction is an essential step in drug repositioning. A few graph neural network (GNN)-based methods have been proposed for DTI using heterogeneous biological data. However, existing GNN-based only aggregate information from directly connected nodes restricted a drug-related or target-related and are incapable of capturing high-order dependencies the graph. In this paper, we propose metapath-aggregated (MHGNN) to capture complex structures rich...

In this paper, we propose a zoom auxiliary imaging lens based on the four-component mechanical method for modulation transfer function (MTF) test system. The lenses of current MTF system typically use fixed-focus optical systems, which are unable to meet scenarios fast and batch measurement measure with an extensive focal length range. Compared lens, can simultaneously satisfy wide-angle telephoto miniature without losing accuracy. entrance pupil distance is greater than that traditional...

The prompt-based learning paradigm has gained much research attention recently. It achieved state-of-the-art performance on several NLP tasks, especially in the few-shot scenarios. While steering downstream few works have been reported to investigate security problems of models. In this paper, we conduct first study vulnerability continuous prompt algorithm backdoor attacks. We observe that scenarios posed a great challenge attacks models, limiting usability existing methods. To address...

Method Rename has been a widely used refactoring operation that improves program comprehension and maintenance. Descriptive method names summarize functionalities of source code can facilitate comprehension. Much research done to suggest through summarization. However, unlike natural language, snippet consists basic blocks organized by complicated structures. In this work, we observe hierarchical structure --- tokens form snippet. Based on observation, exploit attention network learn the...

Identification of drug-target interactions (DTIs) is crucial for drug discovery and repositioning. Existing graph neural network (GNN) based methods only aggregate information from directly connected nodes restricted in a drug-related or target-related network, are incapable capturing long-range dependencies the biological heterogeneous graph. In this paper, we propose attention (HGAN) to capture complex structures rich semantics DTI prediction. HGAN enhances structure learning both...

Extract Method is a widely used refactoring operation to improve method comprehension and maintenance. Much research has been done extract codefragments within the body form new method. Criteria for identifying extractable code usually centered around degrees of cohesiveness, coupling length However, automatic extraction techniques have not highly successful, since it can be hard concretizethe criteria. In this work, we present novel system that learns these criteria refactorings from open...

In recent years, the mortality rate of respiratory diseases ranks high among major diseases. Early detection is a key factor in reducing and curing this paper, we propose BiGRU Attention-XGBoost model to classify sounds, order assist doctors early diagnosis Specifically, first extract two sets features, i.e., time domain spectral features encode sounds. Then, apply Gradient Boosting Decision Tree algorithm select important for classification. Based on temporal characteristics design them....

Open-source software (OSS) licenses dictate the conditions which should be followed to reuse, distribute, and modify software. Apart from widely-used such as MIT License, developers are also allowed customize their own (called custom licenses), whose descriptions more flexible. The presence of various imposes challenges understanding compatibility. To avoid financial legal risks, it is essential ensure license compatibility when integrating third-party packages or reusing code accompanied...

Artificial intelligence (AI) server infrastructure has been built to support AI applications and handle data-intensive workloads. is the essential building blocks, errors in may lead fatal consequences any upon it. Compared traditional software, software for more configurable, thus likely have configuration that might prevent correct behaviors. Previous work on misconfiguration diagnosis requires sufficient execution history or manual intervention, can hardly diagnose potential...

Developers usually use TPLs to facilitate the development of projects avoid reinventing wheels, however, vulnerable indeed cause severe security threats. The majority existing research only considered whether used but neglected code was by projects, which inevitably results in false positives and further requires additional patching efforts maintenance costs. To address this, we propose VAScanner, can effectively identify root methods causing vulnerabilities all APIs Java projects....

Permission mechanisms serve as the main measure to protect users privacy and security in Android applications. Modern smartphone operating systems (Android 6.0 later versions) prompt regulate permissions using ask-on-first-use policy. Much research has been done dynamically depending on user preferences contexts modern operation systems. However, all these techniques have limitations-they heavily rely users' current or historical decisions granting permissions, ignoring fact that are not...

Android API-related compatibility issues have be-come a severe problem and significant challenge for app devel-opers due to the well-known fragmentation issues. To address this problem, many effective approaches such as app-based API lifetime-based methods been proposed identify incompatible usages. However, various implementations of usages different invoking paths, there is still weakness existing approaches, i.e., introducing massive number false positives (FP) negatives (FN). end, in...

One of the major software safety issues is memory leak. Moreover, detecting leak vulnerabilities challenging in static analysis. Existing detection tools find bugs by collecting programs' information process scanning source code. However, current are weak efficiency and accuracy, especially when targeted program contains complex branches. This paper proposes a projection-based approach to detect leaks C code with control flows. According features allocation deallocation code, this projects...

Diffusion models (DMs) have enabled breakthroughs in image synthesis tasks but lack an intuitive interface for consistent image-to-image (I2I) translation. Various methods been explored to address this issue, including mask-based methods, attention-based and image-conditioning. However, it remains a critical challenge enable unpaired I2I translation with pre-trained DMs while maintaining satisfying consistency. This paper introduces Cyclenet, novel simple method that incorporates cycle...

Penetration test is one of the most used SQL Injection Vulnerability (SQLIV) testing technology. Focused on insufficiency accuracy problem in SQLIV black-box penetration process, we discuss limitation traditional approaches based case library enumerating methods and propose a Test approach Finite State Machine (SPT-FSM). The proposed establishes FSM states corresponding to different cases, map statuses cases their relevant responses, analyzes transition regularity established for with...

Among all the Web application security issues, SQL Injection Vulnerability (SQLIV) is one of most serious problems. How to test SQLIV effectively great importance. To address this issue, paper describes a novel approach that utilization Feature Matrix (FM) model for black-box penetration test. Firstly, FM introduced, which integrates general features SQLIV. Each row matrix defined as pattern, named Global Test Pattern (GTP). Then, GTP Selection (GTPS) process used select legal GTPs Secondly,...

Open-source software (OSS) licenses dictate the conditions, which should be followed to reuse, distribute, and modify software. Apart from widely-used such as MIT License, developers are also allowed customize their own (called custom license), whose descriptions more flexible. The presence of various imposes challenges understand compatibility. To avoid financial legal risks, it is essential ensure license compatibility when integrating third-party packages or reusing code accompanied with...