Li Li

Xinyi Hou, Yanjie Zhao, Yue Liu et al.

Large Language Models (LLMs) have significantly impacted numerous domains, including Software Engineering (SE). Many recent publications have explored LLMs applied to various SE tasks. Nevertheless, a comprehensive understanding of the application, effects, and possible limitations of LLMs on SE is still in its early stages. To bridge this gap, we conducted a systematic literature review (SLR) on LLM4SE, with a particular focus on understanding how LLMs can be exploited to optimize processes and outcomes. We select and analyze 395 research papers from January 2017 to January 2024 to answer four key research questions (RQs). In RQ1, we categorize different LLMs that have been employed in SE tasks, characterizing their distinctive features and uses. In RQ2, we analyze the methods used in data collection, preprocessing, and application, highlighting the role of well-curated datasets for successful LLM for SE implementation. RQ3 investigates the strategies employed to optimize and evaluate the performance of LLMs in SE. Finally, RQ4 examines the specific SE tasks where LLMs have shown success to date, illustrating their practical contributions to the field. From the answers to these RQs, we discuss the current state-of-the-art and trends, identifying gaps in existing research, and flagging promising areas for future study. Our artifacts are publicly available at https://github.com/xinyi-hou/LLM4SE_SLR.

Xinyu She, Yue Liu, Yanjie Zhao et al.

Modern language models (LMs) have been successfully employed in source code generation and understanding, leading to a significant increase in research focused on learning-based code intelligence, such as automated bug repair, and test case generation. Despite their great potential, language models for code intelligence (LM4Code) are susceptible to potential pitfalls, which hinder realistic performance and further impact their reliability and applicability in real-world deployment. Such challenges drive the need for a comprehensive understanding - not just identifying these issues but delving into their possible implications and existing solutions to build more reliable language models tailored to code intelligence. Based on a well-defined systematic research approach, we conducted an extensive literature review to uncover the pitfalls inherent in LM4Code. Finally, 67 primary studies from top-tier venues have been identified. After carefully examining these studies, we designed a taxonomy of pitfalls in LM4Code research and conducted a systematic study to summarize the issues, implications, current solutions, and challenges of different pitfalls for LM4Code systems. We developed a comprehensive classification scheme that dissects pitfalls across four crucial aspects: data collection and labeling, system design and learning, performance evaluation, and deployment and maintenance. Through this study, we aim to provide a roadmap for researchers and practitioners, facilitating their understanding and utilization of LM4Code in reliable and trustworthy ways.

Zhibo Zhang, Wuxia Bai, Yuxi Li et al.

Large language models (LLMs) have achieved unprecedented success in the field of natural language processing. However, the black-box nature of their internal mechanisms has brought many concerns about their trustworthiness and interpretability. Recent research has discovered a class of abnormal tokens in the model's vocabulary space and named them "glitch tokens". Those tokens, once included in the input, may induce the model to produce incorrect, irrelevant, or even harmful results, drastically undermining the reliability and practicality of LLMs. In this work, we aim to enhance the understanding of glitch tokens and propose techniques for their detection and mitigation. We first reveal the characteristic features induced by glitch tokens on LLMs, which are evidenced by significant deviations in the distributions of attention patterns and dynamic information from intermediate model layers. Based on the insights, we develop GlitchProber, a tool for efficient glitch token detection and mitigation. GlitchProber utilizes small-scale sampling, principal component analysis for accelerated feature extraction, and a simple classifier for efficient vocabulary screening. Taking one step further, GlitchProber rectifies abnormal model intermediate layer values to mitigate the destructive effects of glitch tokens. Evaluated on five mainstream open-source LLMs, GlitchProber demonstrates higher efficiency, precision, and recall compared to existing approaches, with an average F1 score of 0.86 and an average repair rate of 50.06%. GlitchProber unveils a novel path to address the challenges posed by glitch tokens and inspires future research toward more robust and interpretable LLMs.

Zhensu Sun, Xiaoning Du, Fu Song et al.

Code datasets are of immense value for training neural-network-based code completion models, where companies or organizations have made substantial investments to establish and process these datasets. Unluckily, these datasets, either built for proprietary or public usage, face the high risk of unauthorized exploits, resulting from data leakages, license violations, etc. Even worse, the ``black-box'' nature of neural models sets a high barrier for externals to audit their training datasets, which further connives these unauthorized usages. Currently, watermarking methods have been proposed to prohibit inappropriate usage of image and natural language datasets. However, due to domain specificity, they are not directly applicable to code datasets, leaving the copyright protection of this emerging and important field of code data still exposed to threats. To fill this gap, we propose a method, named CodeMark, to embed user-defined imperceptible watermarks into code datasets to trace their usage in training neural code completion models. CodeMark is based on adaptive semantic-preserving transformations, which preserve the exact functionality of the code data and keep the changes covert against rule-breakers. We implement CodeMark in a toolkit and conduct an extensive evaluation of code completion models. CodeMark is validated to fulfill all desired properties of practical watermarks, including harmlessness to model accuracy, verifiability, robustness, and imperceptibility.

Xin Wang, Yasheng Wang, Yao Wan et al.

Recent years have witnessed increasing interest in code representation learning, which aims to represent the semantics of source code into distributed vectors. Currently, various works have been proposed to represent the complex semantics of source code from different views, including plain text, Abstract Syntax Tree (AST), and several kinds of code graphs (e.g., Control/Data Flow Graph). However, most of them only consider a single view of source code independently, ignoring the correspondences among different views. In this paper, we propose to integrate different views with the natural-language description of source code into a unified framework with Multi-View contrastive Pre-training, and name our model as CODE-MVP. Specifically, we first extract multiple code views using compiler tools, and learn the complementary information among them under a contrastive learning framework. Inspired by the type checking in compilation, we also design a fine-grained type inference objective in the pre-training. Experiments on three downstream tasks over five datasets demonstrate the superiority of CODE-MVP when compared with several state-of-the-art baselines. For example, we achieve 2.4/2.3/1.1 gain in terms of MRR/MAP/Accuracy metrics on natural language code retrieval, code similarity, and code defect detection tasks, respectively.

Wei Hu, Xuhong Wang, Ding Wang et al.

In the realm of software applications in the transportation industry, Domain-Specific Languages (DSLs) have enjoyed widespread adoption due to their ease of use and various other benefits. With the ceaseless progress in computer performance and the rapid development of large-scale models, the possibility of programming using natural language in specified applications - referred to as Application-Specific Natural Language (ASNL) - has emerged. ASNL exhibits greater flexibility and freedom, which, in turn, leads to an increase in computational complexity for parsing and a decrease in processing performance. To tackle this issue, our paper advances a design for an intermediate representation (IR) that caters to ASNL and can uniformly process transportation data into graph data format, improving data processing performance. Experimental comparisons reveal that in standard data query operations, our proposed IR design can achieve a speed improvement of over forty times compared to direct usage of standard XML format data.

Terry Yue Zhuo, Xiaoning Du, Zhenchang Xing et al.

Recent breakthroughs in pre-trained code models, such as CodeBERT and Codex, have shown their superior performance in various downstream tasks. The correctness and unambiguity of API usage among these code models are crucial for achieving desirable program functionalities, requiring them to learn various API fully qualified names structurally and semantically. Recent studies reveal that even state-of-the-art pre-trained code models struggle with suggesting the correct APIs during code generation. However, the reasons for such poor API usage performance are barely investigated. To address this challenge, we propose using knowledge probing as a means of interpreting code models, which uses cloze-style tests to measure the knowledge stored in models. Our comprehensive study examines a code model's capability of understanding API fully qualified names from two different perspectives: API call and API import. Specifically, we reveal that current code models struggle with understanding API names, with pre-training strategies significantly affecting the quality of API name learning. We demonstrate that natural language context can assist code models in locating Python API names and generalize Python API name knowledge to unseen data. Our findings provide insights into the limitations and capabilities of current pre-trained code models, and suggest that incorporating API structure into the pre-training process can improve automated API usage and code representations. This work provides significance for advancing code intelligence practices and direction for future studies. All experiment results, data and source code used in this work are available at \url{https://doi.org/10.5281/zenodo.7902072}.

Jacob Wiebe, Ranwa Al Mallah, Li Li

Recent advancements in deep learning techniques have opened new possibilities for designing solutions for autonomous cyber defence. Teams of intelligent agents in computer network defence roles may reveal promising avenues to safeguard cyber and kinetic assets. In a simulated game environment, agents are evaluated on their ability to jointly mitigate attacker activity in host-based defence scenarios. Defender systems are evaluated against heuristic attackers with the goals of compromising network confidentiality, integrity, and availability. Value-based Independent Learning and Centralized Training Decentralized Execution (CTDE) cooperative Multi-Agent Reinforcement Learning (MARL) methods are compared revealing that both approaches outperform a simple multi-agent heuristic defender. This work demonstrates the ability of cooperative MARL to learn effective cyber defence tactics against varied threats.

Zhihao Lin, Wei Ma, Tao Lin et al.

Large Language Models (LLMs) have become instrumental in advancing software engineering (SE) tasks, showcasing their efficacy in code understanding and beyond. Like traditional SE tools, open-source collaboration is key in realising the excellent products. However, with AI models, the essential need is in data. The collaboration of these AI-based SE models hinges on maximising the sources of high-quality data. However, data especially of high quality, often holds commercial or sensitive value, making it less accessible for open-source AI-based SE projects. This reality presents a significant barrier to the development and enhancement of AI-based SE tools within the software engineering community. Therefore, researchers need to find solutions for enabling open-source AI-based SE models to tap into resources by different organisations. Addressing this challenge, our position paper investigates one solution to facilitate access to diverse organizational resources for open-source AI models, ensuring privacy and commercial sensitivities are respected. We introduce a governance framework centered on federated learning (FL), designed to foster the joint development and maintenance of open-source AI code models while safeguarding data privacy and security. Additionally, we present guidelines for developers on AI-based SE tool collaboration, covering data requirements, model architecture, updating strategies, and version control. Given the significant influence of data characteristics on FL, our research examines the effect of code data heterogeneity on FL performance.

Terry Yue Zhuo, Zhou Yang, Zhensu Sun et al.

The increasingly popular adoption of deep learning models in many critical source code tasks motivates the development of data augmentation (DA) techniques to enhance training data and improve various capabilities (e.g., robustness and generalizability) of these models. Although a series of DA methods have been proposed and tailored for source code models, there lacks a comprehensive survey and examination to understand their effectiveness and implications. This paper fills this gap by conducting a comprehensive and integrative survey of data augmentation for source code, wherein we systematically compile and encapsulate existing literature to provide a comprehensive overview of the field. We start with an introduction of data augmentation in source code and then provide a discussion on major representative approaches. Next, we highlight the general strategies and techniques to optimize the DA quality. Subsequently, we underscore techniques useful in real-world source code scenarios and downstream tasks. Finally, we outline the prevailing challenges and potential opportunities for future research. In essence, we aim to demystify the corpus of existing literature on source code DA for deep learning, and foster further exploration in this sphere. Complementing this, we present a continually updated GitHub repository that hosts a list of update-to-date papers on DA for source code modeling, accessible at \url{https://github.com/terryyz/DataAug4Code}.

Li Li, Jiawei Wang, Haowei Quan

Despite being the most popular programming language, Python has not yet received enough attention from the community. To the best of our knowledge, there is no general static analysis framework proposed to facilitate the implementation of dedicated Python static analyzers. To fill this gap, we design and implement such a framework (named Scalpel) and make it publicly available as an open-source project. The Scalpel framework has already integrated a number of fundamental static analysis functions (e.g., call graph constructions, control-flow graph constructions, alias analysis, etc.) that are ready to be reused by developers to implement client applications focusing on statically resolving dedicated Python problems such as detecting bugs or fixing vulnerabilities.

Zhensu Sun, Xiaoning Du, Fu Song et al.

Github Copilot, trained on billions of lines of public code, has recently become the buzzword in the computer science research and practice community. Although it is designed to help developers implement safe and effective code with powerful intelligence, practitioners and researchers raise concerns about its ethical and security problems, e.g., should the copyleft licensed code be freely leveraged or insecure code be considered for training in the first place? These problems pose a significant impact on Copilot and other similar products that aim to learn knowledge from large-scale open-source code through deep learning models, which are inevitably on the rise with the fast development of artificial intelligence. To mitigate such impacts, we argue that there is a need to invent effective mechanisms for protecting open-source code from being exploited by deep learning models. Here, we design and implement a prototype, CoProtector, which utilizes data poisoning techniques to arm source code repositories for defending against such exploits. Our large-scale experiments empirically show that CoProtector is effective in achieving its purpose, significantly reducing the performance of Copilot-like deep learning models while being able to stably reveal the secretly embedded watermark backdoors.

Xian Zhan, Tianming Liu, Yepang Liu et al.

Third-party libraries (TPLs) have become a significant part of the Android ecosystem. Developers can employ various TPLs to facilitate their app development. Unfortunately, the popularity of TPLs also brings new security issues. For example, TPLs may carry malicious or vulnerable code, which can infect popular apps to pose threats to mobile users. Furthermore, TPL detection is essential for downstream tasks, such as vulnerabilities and malware detection. Thus, various tools have been developed to identify TPLs. However, no existing work has studied these TPL detection tools in detail, and different tools focus on different applications and techniques with performance differences. A comprehensive understanding of these tools will help us make better use of them. To this end, we conduct a comprehensive empirical study to fill the gap by evaluating and comparing all publicly available TPL detection tools based on six criteria: accuracy of TPL construction, effectiveness, efficiency, accuracy of version identification, resiliency to code obfuscation, and ease of use. Besides, we enhance these open-source tools by fixing their limitations, to improve their detection ability. Finally, we build an extensible framework that integrates all existing available TPL detection tools, providing an online service for the research community. We release the evaluation dataset and enhanced tools. According to our study, we also present the essential findings and discuss promising implications to the community. We believe our work provides a clear picture of existing TPL detection techniques and also gives a roadmap for future research.

Kahou Tam, Li Li, Bo Han et al.

Federated learning (FL) collaboratively trains a shared global model depending on multiple local clients, while keeping the training data decentralized in order to preserve data privacy. However, standard FL methods ignore the noisy client issue, which may harm the overall performance of the shared model. We first investigate critical issue caused by noisy clients in FL and quantify the negative impact of the noisy clients in terms of the representations learned by different layers. We have the following two key observations: (1) the noisy clients can severely impact the convergence and performance of the global model in FL, and (2) the noisy clients can induce greater bias in the deeper layers than the former layers of the global model. Based on the above observations, we propose Fed-NCL, a framework that conducts robust federated learning with noisy clients. Specifically, Fed-NCL first identifies the noisy clients through well estimating the data quality and model divergence. Then robust layer-wise aggregation is proposed to adaptively aggregate the local models of each client to deal with the data heterogeneity caused by the noisy clients. We further perform the label correction on the noisy clients to improve the generalization of the global model. Experimental results on various datasets demonstrate that our algorithm boosts the performances of different state-of-the-art systems with noisy clients. Our code is available on https://github.com/TKH666/Fed-NCL

Kunsong Zhao, Jin Liu, Zhou Xu et al.

The software inevitably encounters the crash, which will take developers a large amount of effort to find the fault causing the crash (short for crashing fault). Developing automatic methods to identify the residence of the crashing fault is a crucial activity for software quality assurance. Researchers have proposed methods to predict whether the crashing fault resides in the stack trace based on the features collected from the stack trace and faulty code, aiming at saving the debugging effort for developers. However, previous work usually neglected the feature preprocessing operation towards the crash data and only used traditional classification models. In this paper, we propose a novel crashing fault residence prediction framework, called ConDF, which consists of a consistency based feature subset selection method and a state-of-the-art deep forest model. More specifically, first, the feature selection method is used to obtain an optimal feature subset and reduce the feature dimension by reserving the representative features. Then, a simplified deep forest model is employed to build the classification model on the reduced feature set. The experiments on seven open source software projects show that our ConDF method performs significantly better than 17 baseline methods on three performance indicators.

Mahnush Movahedi, Benjamin M. Case, Andrew Knox et al.

In this paper, we outline a way to deploy a privacy-preserving protocol for multiparty Randomized Controlled Trials on the scale of 500 million rows of data and more than a billion gates. Randomized Controlled Trials (RCTs) are widely used to improve business and policy decisions in various sectors such as healthcare, education, criminology, and marketing. A Randomized Controlled Trial is a scientifically rigorous method to measure the effectiveness of a treatment. This is accomplished by randomly allocating subjects to two or more groups, treating them differently, and then comparing the outcomes across groups. In many scenarios, multiple parties hold different parts of the data for conducting and analyzing RCTs. Given privacy requirements and expectations of each of these parties, it is often challenging to have a centralized store of data to conduct and analyze RCTs. We accomplish this by a three-stage solution. The first stage uses the Private Secret Share Set Intersection (PS$^3$I) solution to create a joined set and establish secret shares without revealing membership, while discarding individuals who were placed into more than one group. The second stage runs multiple instances of a general purpose MPC over a sharded database to aggregate statistics about each experimental group while discarding individuals who took an action before they received treatment. The third stage adds distributed and calibrated Differential Privacy (DP) noise to the aggregate statistics and uncertainty measures, providing formal two-sided privacy guarantees. We also evaluate the performance of multiple open source general purpose MPC libraries for this task. We additionally demonstrate how we have used this to create a working ads effectiveness measurement product capable of measuring hundreds of millions of individuals per experiment.

Daoyuan Li, Li Li, Dongsun Kim et al.

One single code change can significantly influence a wide range of software systems and their users. For example, 1) adding a new feature can spread defects in several modules, while 2) changing an API method can improve the performance of all client programs. Developers often may not clearly know whether their or others' changes are influential at commit time. Rather, it turns out to be influential after affecting many aspects of a system later. This paper investigates influential software changes and proposes an approach to identify them early, i.e., immediately when they are applied. We first conduct a post-mortem analysis to discover existing influential changes by using intuitions such as isolated changes and changes referred by other changes in 10 open source projects. Then we re-categorize all identified changes through an open-card sorting process. Subsequently, we conduct a survey with 89 developers to confirm our influential change categories. Finally, from our ground truth we extract features, including metrics such as the complexity of changes, terms in commit logs and file centrality in co-change graphs, to build machine learning classifiers. The experiment results show that our prediction model achieves overall with random samples 86.8% precision, 74% recall and 80.4% F-measure respectively.

Zhensu Sun, Xiaoning Du, Zhou Yang et al.

Artificial Intelligence (AI) models have emerged as another important audience for programming languages alongside humans and machines, as we enter the era of large language models (LLMs). LLMs can now perform well in coding competitions and even write programs like developers to solve various tasks, including mathematical problems. However, the grammar and layout of current programs are designed to cater the needs of human developers -- with many grammar tokens and formatting tokens being used to make the code easier for humans to read. While this is helpful, such a design adds unnecessary computational work for LLMs, as each token they either use or produce consumes computational resources. To improve inference efficiency and reduce computational costs, we propose the concept of AI-oriented grammar. This aims to represent code in a way that better suits the working mechanism of AI models. Code written with AI-oriented grammar discards formats and uses a minimum number of tokens to convey code semantics effectively. To demonstrate the feasibility of this concept, we explore and implement the first AI-oriented grammar for Python, named SimPy. SimPy is crafted by revising the original Python grammar through a series of heuristic rules. Programs written in SimPy maintain identical AST structures to those in standard Python. This allows for not only execution via a modified AST parser, but also seamless transformation between programs written in Python and SimPy, enabling human developers and LLMs to use Python and SimPy, respectively, when they need to collaborate. In the experiments, compared with Python, SimPy enables a reduction in token usage by 13.5% and 10.4% for CodeLlama and GPT-4, respectively, when completing the same set of code-related tasks. Additionally, these models can maintain or even improve their performance when using SimPy instead of Python for these tasks.

Mingyi Zhou, Xiang Gao, Jing Wu et al.

Numerous mobile apps have leveraged deep learning capabilities. However, on-device models are vulnerable to attacks as they can be easily extracted from their corresponding mobile apps. Existing on-device attacking approaches only generate black-box attacks, which are far less effective and efficient than white-box strategies. This is because mobile deep learning frameworks like TFLite do not support gradient computing, which is necessary for white-box attacking algorithms. Thus, we argue that existing findings may underestimate the harmfulness of on-device attacks. To this end, we conduct a study to answer this research question: Can on-device models be directly attacked via white-box strategies? We first systematically analyze the difficulties of transforming the on-device model to its debuggable version, and propose a Reverse Engineering framework for On-device Models (REOM), which automatically reverses the compiled on-device TFLite model to the debuggable model. Specifically, REOM first transforms compiled on-device models into Open Neural Network Exchange format, then removes the non-debuggable parts, and converts them to the debuggable DL models format that allows attackers to exploit in a white-box setting. Our experimental results show that our approach is effective in achieving automated transformation among 244 TFLite models. Compared with previous attacks using surrogate models, REOM enables attackers to achieve higher attack success rates with a hundred times smaller attack perturbations. In addition, because the ONNX platform has plenty of tools for model format exchanging, the proposed method based on the ONNX platform can be adapted to other model formats. Our findings emphasize the need for developers to carefully consider their model deployment strategies, and use white-box methods to evaluate the vulnerability of on-device models.

Yue Liu, Chakkrit Tantithamthavorn, Li Li

Recent years have witnessed the emerging trend of extensions in modern Integrated Development Environments (IDEs) like Visual Studio Code (VSCode) that significantly enhance developer productivity. Especially, popular AI coding assistants like GitHub Copilot and Tabnine provide conveniences like automated code completion and debugging. While these extensions offer numerous benefits, they may introduce privacy and security concerns to software developers. However, there is no existing work that systematically analyzes the security and privacy concerns, including the risks of data exposure in VSCode extensions. In this paper, we investigate on the security issues of cross-extension interactions in VSCode and shed light on the vulnerabilities caused by data exposure among different extensions. Our study uncovers high-impact security flaws that could allow adversaries to stealthily acquire or manipulate credential-related data (e.g., passwords, API keys, access tokens) from other extensions if not properly handled by extension vendors. To measure their prevalence, we design a novel automated risk detection framework that leverages program analysis and natural language processing techniques to automatically identify potential risks in VSCode extensions. By applying our tool to 27,261 real-world VSCode extensions, we discover that 8.5% of them (i.e., 2,325 extensions) are exposed to credential-related data leakage through various vectors, such as commands, user input, and configurations. Our study sheds light on the security challenges and flaws of the extension-in-IDE paradigm and provides suggestions and recommendations for improving the security of VSCode extensions and mitigating the risks of data exposure.

Zhisheng Yang, Xiaofei Xu, Ke Deng et al.

As powerful tools in Natural Language Processing (NLP), Large Language Models (LLMs) have been leveraged for crafting recommendations to achieve precise alignment with user preferences and elevate the quality of the recommendations. The existing approaches implement both non-tuning and tuning strategies. Compared to following the tuning strategy, the approaches following the non-tuning strategy avoid the relatively costly, time-consuming, and expertise-requiring process of further training pre-trained LLMs on task-specific datasets, but they suffer the issue of not having the task-specific business or local enterprise knowledge. To the best of our knowledge, none of the existing approaches following the non-tuning strategy explicitly integrates collaborative filtering, one of the most successful recommendation techniques. This study aims to fill the gap by proposing critique-based LLMs as recommendation systems (Critic-LLM-RS). For our purpose, we train a separate machine-learning model called Critic that implements collaborative filtering for recommendations by learning from the interactions between many users and items. The Critic provides critiques to LLMs to significantly refine the recommendations. Extensive experiments have verified the effectiveness of Critic-LLM-RS on real datasets.

Zhensu Sun, Li Li, Yan Liu et al.

The performance of neural code search is significantly influenced by the quality of the training data from which the neural models are derived. A large corpus of high-quality query and code pairs is demanded to establish a precise mapping from the natural language to the programming language. Due to the limited availability, most widely-used code search datasets are established with compromise, such as using code comments as a replacement of queries. Our empirical study on a famous code search dataset reveals that over one-third of its queries contain noises that make them deviate from natural user queries. Models trained through noisy data are faced with severe performance degradation when applied in real-world scenarios. To improve the dataset quality and make the queries of its samples semantically identical to real user queries is critical for the practical usability of neural code search. In this paper, we propose a data cleaning framework consisting of two subsequent filters: a rule-based syntactic filter and a model-based semantic filter. This is the first framework that applies semantic query cleaning to code search datasets. Experimentally, we evaluated the effectiveness of our framework on two widely-used code search models and three manually-annotated code retrieval benchmarks. Training the popular DeepCS model with the filtered dataset from our framework improves its performance by 19.2% MRR and 21.3% Answer@1, on average with the three validation benchmarks.

Xiaoyu Sun, Xiao Chen, Kui Liu et al.

While extremely valuable to achieve advanced functions, mobile phone sensors can be abused by attackers to implement malicious activities in Android apps, as experimentally demonstrated by many state-of-the-art studies. There is hence a strong need to regulate the usage of mobile sensors so as to keep them from being exploited by malicious attackers. However, despite the fact that various efforts have been put in achieving this, i.e., detecting privacy leaks in Android apps, we have not yet found approaches to automatically detect sensor leaks in Android apps. To fill the gap, we designed and implemented a novel prototype tool, SEEKER, that extends the famous FlowDroid tool to detect sensor-based data leaks in Android apps. SEEKER conducts sensor-focused static taint analyses directly on the Android apps' bytecode and reports not only sensor-triggered privacy leaks but also the sensor types involved in the leaks. Experimental results using over 40,000 real-world Android apps show that SEEKER is effective in detecting sensor leaks in Android apps, and malicious apps are more interested in leaking sensor data than benign apps.

Jordan Samhi, Li Li, Tegawendé F. Bissyandé et al.

One prominent tactic used to keep malicious behavior from being detected during dynamic test campaigns is logic bombs, where malicious operations are triggered only when specific conditions are satisfied. Defusing logic bombs remains an unsolved problem in the literature. In this work, we propose to investigate Suspicious Hidden Sensitive Operations (SHSOs) as a step towards triaging logic bombs. To that end, we develop a novel hybrid approach that combines static analysis and anomaly detection techniques to uncover SHSOs, which we predict as likely implementations of logic bombs. Concretely, Difuzer identifies SHSO entry-points using an instrumentation engine and an inter-procedural data-flow analysis. Then, it extracts trigger-specific features to characterize SHSOs and leverages One-Class SVM to implement an unsupervised learning model for detecting abnormal triggers. We evaluate our prototype and show that it yields a precision of 99.02% to detect SHSOs among which 29.7% are logic bombs. Difuzer outperforms the state-of-the-art in revealing more logic bombs while yielding less false positives in about one order of magnitude less time. All our artifacts are released to the community.

Roland Croft, M. Ali Babar, Li Li

Software Vulnerability (SV) severity assessment is a vital task for informing SV remediation and triage. Ranking of SV severity scores is often used to advise prioritization of patching efforts. However, severity assessment is a difficult and subjective manual task that relies on expertise, knowledge, and standardized reporting schemes. Consequently, different data sources that perform independent analysis may provide conflicting severity rankings. Inconsistency across these data sources affects the reliability of severity assessment data, and can consequently impact SV prioritization and fixing. In this study, we investigate severity ranking inconsistencies over the SV reporting lifecycle. Our analysis helps characterize the nature of this problem, identify correlated factors, and determine the impacts of inconsistency on downstream tasks. Our findings observe that SV severity often lacks consideration or is underestimated during initial reporting, and such SVs consequently receive lower prioritization. We identify six potential attributes that are correlated to this misjudgment, and show that inconsistency in severity reporting schemes can severely degrade the performance of downstream severity prediction by up to 77%. Our findings help raise awareness of SV severity data inconsistencies and draw attention to this data quality problem. These insights can help developers better consider SV severity data sources, and improve the reliability of consequent SV prioritization. Furthermore, we encourage researchers to provide more attention to SV severity data selection.

Shangwen Wang, Kui Liu, Bo Lin et al.

Software Fault Localization refers to the activity of finding code elements (e.g., statements) that are related to a software failure. The state-of-the-art fault localization techniques, however, produce coarse-grained results that can deter manual debugging or mislead automated repair tools. In this work, we focus specifically on the fine-grained identification of code elements (i.e., tokens) that must be changed to fix a buggy program: we refer to it as fix localization. This paper introduces a neural network architecture (named Beep) that builds on AST paths to predict the buggy code element as well as the change action that must be applied to repair a program. Leveraging massive data of bugs and patches within the CoCoNut dataset, we trained a model that was (1) effective in localizing the buggy tokens with the Mean First Rank significantly higher than a statistics based baseline and a machine learning-based baseline, and (2) effective in predicting the repair operators (with the associated buggy code elements) with a Recall@1= 30-45% and the Mean First Rank=7-12 (evaluated by CoCoNut, ManySStuBs4J, and Defects4J datasets). To showcase how fine-grained fix localization can help program repair, we employ it in two repair pipelines where we use either a code completion engine to predict the correct token or a set of heuristics to search for the suitable donor code. A key strength of accurate fix localization for program repair is that it reduces the chance of patch overfitting, a challenge in generate-and-validate automated program repair: both two repair pipelines achieve a correctness ratio of 100%, i.e., all generated patches are found to be correct. Moreover, accurate fix localization helps enhance the efficiency of program repair.

Humphrey Obie, Mojtaba Shahin, John Grundy et al.

The elicitation of end-users' human values - such as freedom, honesty, transparency, etc. - is important in the development of software systems. We carried out two preliminary Q-studies to understand (a) the general human value opinion types of eHealth applications (apps) end-users (b) the eHealth domain human value opinion types of eHealth apps end-users (c) whether there are differences between the general and eHealth domain opinion types. Our early results show three value opinion types using generic value instruments: (1) fun-loving, success-driven and independent end-user, (2) security-conscious, socially-concerned, and success-driven end-user, and (3) benevolent, success-driven, and conformist end-user Our results also show two value opinion types using domain-specific value instruments: (1) security-conscious, reputable, and honest end-user, and (2) success-driven, reputable and pain-avoiding end-user. Given these results, consideration should be given to domain context in the design and application of values elicitation instruments.

Mojtaba Shahin, Chris Gonsalvez, Jon Whittle et al.

Computational Thinking (CT) has been investigated from different perspectives. This research aims to investigate how secondary school girls perceive CT practices -- the problem-solving practices that students apply while they are engaged in programming -- when using the micro:bit device in a collaborative setting. This study also explores the collaborative programming process of secondary school girls with the micro:bit device. We conducted mixed-methods research with 203 secondary school girls (in the state of Victoria, Australia) and 31 mentors attending a girls-only CT program (OzGirlsCT program). The girls were grouped into 52 teams and collaboratively developed computational solutions around realistic, important problems to them and their communities. We distributed two surveys (with 193 responses each) to the girls. Further, we surveyed the mentors (with 31 responses) who monitored the girls, and collected their observation reports on their teams. Our study indicates that the girls found "debugging" the most difficult type of CT practice to apply, while collaborative practices of CT were the easiest. We found that prior coding experience significantly reduced the difficulty level of only one CT practice - "debugging". Our study also identified six challenges the girls faced and six best practices they adopted when working on their computational solutions.

Li Li, Raed Fayad, Adrian Taylor

Given the success of reinforcement learning (RL) in various domains, it is promising to explore the application of its methods to the development of intelligent and autonomous cyber agents. Enabling this development requires a representative RL training environment. To that end, this work presents CyGIL: an experimental testbed of an emulated RL training environment for network cyber operations. CyGIL uses a stateless environment architecture and incorporates the MITRE ATT&CK framework to establish a high fidelity training environment, while presenting a sufficiently abstracted interface to enable RL training. Its comprehensive action space and flexible game design allow the agent training to focus on particular advanced persistent threat (APT) profiles, and to incorporate a broad range of potential threats and vulnerabilities. By striking a balance between fidelity and simplicity, it aims to leverage state of the art RL algorithms for application to real-world cyber defence.

Xin Wang, Yasheng Wang, Fei Mi et al.

Code representation learning, which aims to encode the semantics of source code into distributed vectors, plays an important role in recent deep-learning-based models for code intelligence. Recently, many pre-trained language models for source code (e.g., CuBERT and CodeBERT) have been proposed to model the context of code and serve as a basis for downstream code intelligence tasks such as code search, code clone detection, and program translation. Current approaches typically consider the source code as a plain sequence of tokens, or inject the structure information (e.g., AST and data-flow) into the sequential model pre-training. To further explore the properties of programming languages, this paper proposes SynCoBERT, a syntax-guided multi-modal contrastive pre-training approach for better code representations. Specially, we design two novel pre-training objectives originating from the symbolic and syntactic properties of source code, i.e., Identifier Prediction (IP) and AST Edge Prediction (TEP), which are designed to predict identifiers, and edges between two nodes of AST, respectively. Meanwhile, to exploit the complementary information in semantically equivalent modalities (i.e., code, comment, AST) of the code, we propose a multi-modal contrastive learning strategy to maximize the mutual information among different modalities. Extensive experiments on four downstream tasks related to code intelligence show that SynCoBERT advances the state-of-the-art with the same pre-training corpus and model size.

Xian Zhan, Tianming Liu, Lingling Fan et al.

Third-party libraries (TPLs) have been widely used in mobile apps, which play an essential part in the entire Android ecosystem. However, TPL is a double-edged sword. On the one hand, it can ease the development of mobile apps. On the other hand, it also brings security risks such as privacy leaks or increased attack surfaces (e.g., by introducing over-privileged permissions) to mobile apps. Although there are already many studies for characterizing third-party libraries, including automated detection, security and privacy analysis of TPLs, TPL attributes analysis, etc., what strikes us odd is that there is no systematic study to summarize those studies' endeavors. To this end, we conduct the first systematic literature review on Android TPL-related research. Following a well-defined systematic literature review protocol, we collected 74 primary research papers closely related to the Android third-party library from 2012 to 2020. After carefully examining these studies, we designed a taxonomy of TPL-related research studies and conducted a systematic study to summarize current solutions, limitations, challenges and possible implications of new research directions related to third-party library analysis. We hope that these contributions can give readers a clear overview of existing TPL-related studies and inspire them to go beyond the current status quo by advancing the discipline with innovative approaches.

Xiaolei Ren, Michael Ho, Jiang Ming et al.

Since compiler optimization is the most common source contributing to binary code differences in syntax, testing the resilience against the changes caused by different compiler optimization settings has become a standard evaluation step for most binary diffing approaches. For example, 47 top-venue papers in the last 12 years compared different program versions compiled by default optimization levels (e.g., -Ox in GCC and LLVM). Although many of them claim they are immune to compiler transformations, it is yet unclear about their resistance to non-default optimization settings. Especially, we have observed that adversaries explored non-default compiler settings to amplify malware differences. This paper takes the first step to systematically studying the effectiveness of compiler optimization on binary code differences. We tailor search-based iterative compilation for the auto-tuning of binary code differences. We develop BinTuner to search near-optimal optimization sequences that can maximize the amount of binary code differences. We run BinTuner with GCC 10.2 and LLVM 11.0 on SPEC benchmarks (CPU2006 & CPU2017), Coreutils, and OpenSSL. Our experiments show that at the cost of 279 to 1,881 compilation iterations, BinTuner can find custom optimization sequences that are substantially better than the general -Ox settings. BinTuner's outputs seriously undermine prominent binary diffing tools' comparisons. In addition, the detection rate of the IoT malware variants tuned by BinTuner falls by more than 50%. Our findings paint a cautionary tale for security analysts that attackers have a new way to mutate malware code cost-effectively, and the research community needs to step back to reassess optimization-resistance evaluations.

Md. Shamsujjoha, John Grundy, Li Li et al.

Classifying mobile apps based on their description is beneficial for several purposes. However, many app descriptions do not reflect app functionalities, whether accidentally or on purpose. Most importantly, these app classification methods do not work if the app description is unavailable. This paper investigates a Reverse Engineering-based Approach to Classify mobile apps using The data that exists in the app, called REACT. To validate the proposed REACT method, we use a large set of Android apps (24,652 apps in total). We also show REACTs' extendibility for malware/anomaly detection and prove its reliability and scalability. However, our analysis shows some limitations in REACT procedure and implementation, especially for similar feature based app grouping. We discuss the root cause of these failures, our key lessons learned, and some future enhancement ideas. We also share our REACT tools and reproduced datasets for the app market analyst, mobile app developers and software engineering research communities for further research purposes.

Yue Liu, Chakkrit Tantithamthavorn, Li Li et al.

Malicious applications (particularly those targeting the Android platform) pose a serious threat to developers and end-users. Numerous research efforts have been devoted to developing effective approaches to defend against Android malware. However, given the explosive growth of Android malware and the continuous advancement of malicious evasion technologies like obfuscation and reflection, Android malware defense approaches based on manual rules or traditional machine learning may not be effective. In recent years, a dominant research field called deep learning (DL), which provides a powerful feature abstraction ability, has demonstrated a compelling and promising performance in a variety of areas, like natural language processing and computer vision. To this end, employing deep learning techniques to thwart Android malware attacks has recently garnered considerable research attention. Yet, no systematic literature review focusing on deep learning approaches for Android Malware defenses exists. In this paper, we conducted a systematic literature review to search and analyze how deep learning approaches have been applied in the context of malware defenses in the Android environment. As a result, a total of 132 studies covering the period 2014-2021 were identified. Our investigation reveals that, while the majority of these sources mainly consider DL-based on Android malware detection, 53 primary studies (40.1 percent) design defense approaches based on other scenarios. This review also discusses research trends, research focuses, challenges, and future research directions in DL-based Android malware defenses.

Jiawei Wang, Li Li, Andreas Zeller

More than ninety percent of published Jupyter notebooks do not state dependencies on external packages. This makes them non-executable and thus hinders reproducibility of scientific results. We present SnifferDog, an approach that 1) collects the APIs of Python packages and versions, creating a database of APIs; 2) analyzes notebooks to determine candidates for required packages and versions; and 3) checks which packages are required to make the notebook executable (and ideally, reproduce its stored results). In its evaluation, we show that SnifferDog precisely restores execution environments for the largest majority of notebooks, making them immediately executable for end users.

Rui Zhang, Kai Yin, Li Li

The performance of a task-completion dialogue agent usually affects the user experience: when the conversation system yields an unreasonable response, users may feel dissatisfied. Besides, early termination often occurs in disappointing conversations. However, existing off-the-shelf user simulators generally assume an ideal and cooperative user, which is somewhat different from a real user, and inevitably lead to a sub-optimal dialogue policy. In this paper, we propose an emotion-aware user simulation framework for task-oriented dialogue, which is based on the OCC emotion model to update user emotions and drive user actions, to generate simulated behaviors that more similar to real users. We present a linear implementation (The source code will be released soon.) that is easy to understand and extend, and evaluate it on two domain-specific datasets. The experimental results show that the emotional simulation results of our proposed framework conform to common sense and have good versatility for different domains. Meanwhile, our framework provides us with another perspective to understand the improvement process of the dialogue policy model based on reinforcement learning.

Haoye Tian, Kui Liu, Abdoul Kader Kaboreé et al.

A large body of the literature of automated program repair develops approaches where patches are generated to be validated against an oracle (e.g., a test suite). Because such an oracle can be imperfect, the generated patches, although validated by the oracle, may actually be incorrect. While the state of the art explore research directions that require dynamic information or rely on manually-crafted heuristics, we study the benefit of learning code representations to learn deep features that may encode the properties of patch correctness. Our work mainly investigates different representation learning approaches for code changes to derive embeddings that are amenable to similarity computations. We report on findings based on embeddings produced by pre-trained and re-trained neural networks. Experimental results demonstrate the potential of embeddings to empower learning algorithms in reasoning about patch correctness: a machine learning predictor with BERT transformer-based embeddings associated with logistic regression yielded an AUC value of about 0.8 in predicting patch correctness on a deduplicated dataset of 1000 labeled patches. Our study shows that learned representations can lead to reasonable performance when comparing against the state-of-the-art, PATCH-SIM, which relies on dynamic information. These representations may further be complementary to features that were carefully (manually) engineered in the literature.

Pingfan Kong, Li Li, Jun Gao et al.

Android framework-specific app crashes are hard to debug. Indeed, the callback-based event-driven mechanism of Android challenges crash localization techniques that are developed for traditional Java programs. The key challenge stems from the fact that the buggy code location may not even be listed within the stack trace. For example, our empirical study on 500 framework-specific crashes from an open benchmark has revealed that 37 percent of the crash types are related to bugs that are outside the stack traces. Moreover, Android programs are a mixture of code and extra-code artifacts such as the Manifest file. The fact that any artifact can lead to failures in the app execution creates the need to position the localization target beyond the code realm. In this paper, we propose Anchor, a two-phase suspicious bug location suggestion tool. Anchor specializes in finding crash-inducing bugs outside the stack trace. Anchor is lightweight and source code independent since it only requires the crash message and the apk file to locate the fault. Experimental results, collected via cross-validation and in-the-wild dataset evaluation, show that Anchor is effective in locating Android framework-specific crashing faults.

Xi Liu, Li Li, Ping-Chun Hsieh et al.

With the explosive growth of online products and content, recommendation techniques have been considered as an effective tool to overcome information overload, improve user experience, and boost business revenue. In recent years, we have observed a new desideratum of considering long-term rewards of multiple related recommendation tasks simultaneously. The consideration of long-term rewards is strongly tied to business revenue and growth. Learning multiple tasks simultaneously could generally improve the performance of individual task due to knowledge sharing in multi-task learning. While a few existing works have studied long-term rewards in recommendations, they mainly focus on a single recommendation task. In this paper, we propose {\it PoDiRe}: a \underline{po}licy \underline{di}stilled \underline{re}commender that can address long-term rewards of recommendations and simultaneously handle multiple recommendation tasks. This novel recommendation solution is based on a marriage of deep reinforcement learning and knowledge distillation techniques, which is able to establish knowledge sharing among different tasks and reduce the size of a learning model. The resulting model is expected to attain better performance and lower response latency for real-time recommendation services. In collaboration with Samsung Game Launcher, one of the world's largest commercial mobile game platforms, we conduct a comprehensive experimental study on large-scale real data with hundreds of millions of events and show that our solution outperforms many state-of-the-art methods in terms of several standard evaluation metrics.

Arthur D. Sawadogo, Tegawendé F. Bissyandé, Naouel Moha et al.

Timely patching is paramount to safeguard users and maintainers against dire consequences of malicious attacks. In practice, patching is prioritized following the nature of the code change that is committed in the code repository. When such a change is labeled as being security-relevant, i.e., as fixing a vulnerability, maintainers rapidly spread the change and users are notified about the need to update to a new version of the library or of the application. Unfortunately, oftentimes, some security-relevant changes go unnoticed as they represent silent fixes of vulnerabilities. In this paper, we propose a Co-Training-based approach to catch security patches as part of an automatic monitoring service of code repositories. Leveraging different classes of features, we empirically show that such automation is feasible and can yield a precision of over 90% in identifying security patches, with an unprecedented recall of over 80%. Beyond such a benchmarking with ground truth data which demonstrates an improvement over the state-of-the-art, we confirmed that our approach can help catch security patches that were not reported as such.

Luis Cruz, Rui Abreu, John Grundy et al.

Energy efficiency is a crucial quality requirement for mobile applications. However, improving energy efficiency is far from trivial as developers lack the knowledge and tools to aid in this activity. In this paper we study the impact of changes to improve energy efficiency on the maintainability of Android applications. Using a dataset containing 539 energy efficiency-oriented commits, we measure maintainability -- as computed by the Software Improvement Group's web-based source code analysis service Better Code Hub (BCH) -- before and after energy efficiency-related code changes. Results show that in general improving energy efficiency comes with a significant decrease in maintainability. This is particularly evident in code changes to accommodate the Power Save Mode and Wakelock Addition energy patterns. In addition, we perform manual analysis to assess how real examples of energy-oriented changes affect maintainability. Our results help mobile app developers to 1) avoid common maintainability issues when improving the energy efficiency of their apps; and 2) adopt development processes to build maintainable and energy-efficient code. We also support researchers by identifying challenges in mobile app development that still need to be addressed.

Wentao Ouyang, Xiuwu Zhang, Shukui Ren et al.

Click-through rate (CTR) prediction is a critical task in online advertising systems. Models like Deep Neural Networks (DNNs) are simple but stateless. They consider each target ad independently and cannot directly extract useful information contained in users' historical ad impressions and clicks. In contrast, models like Recurrent Neural Networks (RNNs) are stateful but complex. They model temporal dependency between users' sequential behaviors and can achieve improved prediction performance than DNNs. However, both the offline training and online prediction process of RNNs are much more complex and time-consuming. In this paper, we propose Memory Augmented DNN (MA-DNN) for practical CTR prediction services. In particular, we create two external memory vectors for each user, memorizing high-level abstractions of what a user possibly likes and dislikes. The proposed MA-DNN achieves a good compromise between DNN and RNN. It is as simple as DNN, but has certain ability to exploit useful information contained in users' historical behaviors as RNN. Both offline and online experiments demonstrate the effectiveness of MA-DNN for practical CTR prediction services. Actually, the memory component can be augmented to other models as well (e.g., the Wide&Deep model).

Jiawei Wang, Li Li, Andreas Zeller

By bringing together code, text, and examples, Jupyter notebooks have become one of the most popular means to produce scientific results in a productive and reproducible way. As many of the notebook authors are experts in their scientific fields, but laymen with respect to software engineering, one may ask questions on the quality of notebooks and their code. In a preliminary study, we experimentally demonstrate that Jupyter notebooks are inundated with poor quality code, e.g., not respecting recommended coding practices, or containing unused variables and deprecated functions. Considering the education nature of Jupyter notebooks, these poor coding practices as well as the lacks of quality control might be propagated into the next generation of developers. Hence, we argue that there is a strong need to programmatically analyze Jupyter notebooks, calling on our community to pay more attention to the reliability of Jupyter notebooks.

Ayesha Sadiq, Yuan-Fang Li, Li Li et al.

In mainstream programming languages such as Java, a common way to enable concurrency is to manually introduce explicit concurrency constructs such as multi-threading. In multi-threaded programs, managing synchronization between threads is a complicated and challenging task for the programmers due to thread interleaving and heap interference that leads to problems such as deadlocks, data races. With these considerations in mind, access permission-based dependencies have been investigated as an alternative approach to verify the correctness of multi-threaded programs and to exploit the implicit concurrency present in sequential programs without using explicit concurrency constraints. However, significant annotation overhead can arise from manually adding permission-based specifications in a source program, diminishing the effectiveness of existing permission-based approaches. In this paper,we present a framework, Sip4J, to automatically extract access permission-based implicit dependencies from sequential Java programs, by performing inter-procedural static analysis of the source code. Moreover, we integrate and extend an existing permission-based verification tool, Pulse, to automatically verify correctness of the inferred specifications and to reason about their concurrent behaviors. Our evaluation on some widely-used benchmarks gives strong evidence of the correctness of the inferred annotations and their effectiveness in enabling concurrency in sequential programs.

Li Li, Tegawendé Bissyandé, Jacques Klein

Repackaging is a serious threat to the Android ecosystem as it deprives app developers of their benefits, contributes to spreading malware on users' devices, and increases the workload of market maintainers. In the space of six years, the research around this specific issue has produced 57 approaches which do not readily scale to millions of apps or are only evaluated on private datasets without, in general, tool support available to the community. Through a systematic literature review of the subject, we argue that the research is slowing down, where many state-of-the-art approaches have reported high-performance rates on closed datasets, which are unfortunately difficult to replicate and to compare against. In this work, we propose to reboot the research in repackaged app detection by providing a literature review that summarises the challenges and current solutions for detecting repackaged apps and by providing a large dataset that supports replications of existing solutions and implications of new research directions. We hope that these contributions will re-activate the direction of detecting repackaged apps and spark innovative approaches going beyond the current state-of-the-art.

Haoyu Wang, Zhe Liu, Jingyue Liang et al.

China is one of the largest Android markets in the world. As Chinese users cannot access Google Play to buy and install Android apps, a number of independent app stores have emerged and compete in the Chinese app market. Some of the Chinese app stores are pre-installed vendor-specific app markets (e.g., Huawei, Xiaomi and OPPO), whereas others are maintained by large tech companies (e.g., Baidu, Qihoo 360 and Tencent). The nature of these app stores and the content available through them vary greatly, including their trustworthiness and security guarantees. As of today, the research community has not studied the Chinese Android ecosystem in depth. To fill this gap, we present the first large-scale comparative study that covers more than 6 million Android apps downloaded from 16 Chinese app markets and Google Play. We focus our study on catalog similarity across app stores, their features, publishing dynamics, and the prevalence of various forms of misbehavior (including the presence of fake, cloned and malicious apps). Our findings also suggest heterogeneous developer behavior across app stores, in terms of code maintenance, use of third-party services, and so forth. Overall, Chinese app markets perform substantially worse when taking active measures to protect mobile users and legit developers from deceptive and abusive actors, showing a significantly higher prevalence of malware, fake, and cloned apps than Google Play.

Yangyu Hu, Haoyu Wang, Yajin Zhou et al.

In this work, we are focusing on a new and yet uncovered way for malicious apps to gain profit. They claim to be dating apps. However, their sole purpose is to lure users into purchasing premium/VIP services to start conversations with other (likely fake female) accounts in the app. We call these apps as fraudulent dating apps. This paper performs a systematic study to understand the whole ecosystem of fraudulent dating apps. Specifically, we have proposed a three-phase method to detect them and subsequently comprehend their characteristics via analyzing the existing account profiles. Our observation reveals that most of the accounts are not managed by real persons, but by chatbots based on predefined conversation templates. We also analyze the business model of these apps and reveal that multiple parties are actually involved in the ecosystem, including producers who develop apps, publishers who publish apps to gain profit, and the distribution network that is responsible for distributing apps to end users. Finally, we analyze the impact of them to users (i.e., victims) and estimate the overall revenue. Our work is the first systematic study on fraudulent dating apps, and the results demonstrate the urge for a solution to protect users.

Lei Ma, Fuyuan Zhang, Jiyuan Sun et al.

Deep learning (DL) defines a new data-driven programming paradigm where the internal system logic is largely shaped by the training data. The standard way of evaluating DL models is to examine their performance on a test dataset. The quality of the test dataset is of great importance to gain confidence of the trained models. Using an inadequate test dataset, DL models that have achieved high test accuracy may still lack generality and robustness. In traditional software testing, mutation testing is a well-established technique for quality evaluation of test suites, which analyzes to what extent a test suite detects the injected faults. However, due to the fundamental difference between traditional software and deep learning-based software, traditional mutation testing techniques cannot be directly applied to DL systems. In this paper, we propose a mutation testing framework specialized for DL systems to measure the quality of test data. To do this, by sharing the same spirit of mutation testing in traditional software, we first define a set of source-level mutation operators to inject faults to the source of DL (i.e., training data and training programs). Then we design a set of model-level mutation operators that directly inject faults into DL models without a training process. Eventually, the quality of test data could be evaluated from the analysis on to what extent the injected faults could be detected. The usefulness of the proposed mutation testing techniques is demonstrated on two public datasets, namely MNIST and CIFAR-10, with three DL models.

Pedro Silva, Sepehr Akhavan-Masouleh, Li Li

Detecting PE malware files is now commonly approached using statistical and machine learning models. While these models commonly use features extracted from the structure of PE files, we propose that icons from these files can also help better predict malware. We propose an innovative machine learning approach to extract information from icons. Our proposed approach consists of two steps: 1) extracting icon features using summary statics, histogram of gradients (HOG), and a convolutional autoencoder, 2) clustering icons based on the extracted icon features. Using publicly available data and by using machine learning experiments, we show our proposed icon clusters significantly boost the efficacy of malware prediction models. In particular, our experiments show an average accuracy increase of 10% when icon clusters are used in the prediction model.

Li Li, Jun Gao, Médéric Hurier et al.

We present a growing collection of Android apps collected from several sources, including the official Google Play app market and a growing collection of various metadata of those collected apps aiming at facilitating the Android-relevant research works. Our dataset by far has collected over five million apps and over 20 types of metadata such as VirusTotal reports. Our objective of collecting this dataset is to contribute to ongoing research efforts, as well as to enable new potential research topics on Android Apps. By releasing our app and metadata set to the research community, we also aim at encouraging our fellow researchers to engage in reproducible experiments. This article will be continuously updated based on the growing apps and metadata collected in the AndroZoo project. If you have specific metadata that you want to collect from AndroZoo and which are not yet provided by far, please let us know. We will thereby prioritise it in our collecting process so as to provide it to our fellow researchers in a short manner.