Abstract As semiconductor manufacturing becomes increasingly outsourced to untrusted entities, Hardware Trojan (HT) attacks pose a critical threat to the security and reliability of modern integrated circuits. Machine learning models have improved the effectiveness of HT detection using Ring Oscillator Network (RON) side-channel data, yet recent work shows that these models are highly vulnerable to adversarial attacks. This paper evaluates the robustness of the Support Vector Machine (SVM) classifier, a leading algorithm in state-of-the-art HT detection frameworks, under gradient-based adversarial attacks. The proposed work demonstrates that high nominal accuracy does not ensure security against these attacks, which can reduce recall to zero. To strengthen resilience, three data-augmentation methods are investigated: SMOTE, Conditional Tabular Generative Adversarial Network (CTGAN), and Tabular Variational Autoencoder (TVAE). TVAE produces high-fidelity synthetic samples and substantially improves robustness, maintaining over 91% accuracy for nominal performance and over 88% accuracy under strong adversarial perturbations that cause a 100% attack success rate for the surrogate model. The results highlight the need to reframe hardware security evaluations beyond nominal accuracy toward adversarial robustness.

Detecting software vulnerabilities is crucial for securing cyberspace. Traditional learning-based vulnerability detection methods rely on training models from scratch or fine-tuning pre-trained models, both of which require large labeled datasets and significant computational resources. However, acquiring well-annotated vulnerability datasets is often impractical due to high verification costs and data scarcity. To address this challenge, we propose an innovative system, RearVul , which Re parameterizes a dversarial r eprogramming in a low-dimensional subspace for software vul nerability detection. Unlike conventional approaches, RearVul repurposes a pre-trained image classification model using adversarial reprogramming, enabling vulnerability detection with minimal modifications to the model. By learning a universal perturbation applied to program representations, our approach retains the original model’s feature extraction capabilities while adapting it to a new domain. Furthermore, we introduce a low-dimensional reparameterization strategy that decomposes the perturbation into two smaller matrices and optimizes them during training, significantly reducing the number of trainable parameters and computational overhead. Extensive evaluations on publicly available vulnerability datasets demonstrate that RearVul achieves competitive detection accuracy while drastically improving efficiency. Notably, RearVul reduces end-to-end training time to 75min, while maintaining competitive accuracy. In data-limited settings with only 100 training samples, RearVul achieves the best accuracy (65.64%), consistently outperforming recent state-of-the-art baselines and transfer learning via fine-tuning ImageNet backbones. These results highlight its adaptability and effectiveness for real-world software vulnerability detection.

Abstract Malicious bots typically operate within networks through peer-to-peer (P2P) communication structures, leading to the emergence of graph neural networks (GNNs) as a promising bot detection method. However, communications graphs representing bot-infected networks often exhibit an inherent imbalance, coupled with a high degree of heterophily. Graph oversampling techniques, employed to address class imbalance on graphs, are burdened with downsides, such as the creation of complex and noisy topological structures or further amplification of heterophily in a graph. Out-of-distribution detection (ODD) is considered as an alternative solution to address data imbalance issues, but when applied to graphs, this belief is built on an assumption that the underlying graph structure does not interfere with the learning of data distributions. In this paper, we propose a new ODD model HistNet which implements H eterophily-aware is o t ropic out-of-distribution detection to explore how to leverage ODD for malicious bot detection in a Net work. HistNet proceeds with heterophily-aware node embedding that facilitates enhanced isotropic distance calculation and homophily-augmented distance-based belief propagation, which is further regularized by implicit clustering. These technical designs enable HistNet to overcome performance issues caused by imbalance and heterophily in graphs and improve isotropic ODD for bot detection. We validate our claims through extensive experiments on 10 computer networks derived from TON IoT datasets, which comprise real captured bot data. The experimental results demonstrate that HistNet achieves state-of-the-art performance in malicious bot detection on graphs with high graph heterophily and extreme class imbalance.

Phishing emails are an escalating threat, underscoring the need for precise detection methods. While large language models (LLMs) have gained attention for their potential in this area, their reliance on extensive data for fine-tuning poses practical challenges. This paper introduces DualLM for phishing detection with minimal data, which distills the reasoning ability from a large LM to enhance a small target LM and integrates trainable perturbations to improve the small LM's inference capabilities. Experiments demonstrate that DualLM can benefit from dual LMs, which reduces training parameters and data required, while maintaining high performance in phishing email detection with limited data.

Large Language Models (LLMs) have demonstrated significant potential across various applications, but their use as AI copilots in complex and specialized tasks is often hindered by AI hallucinations, where models generate outputs that seem plausible but are incorrect. To address this challenge, we develop AutoFEA, an intelligent system that integrates LLMs with Finite Element Analysis (FEA) to automate the generation of FEA input files. Our approach features a novel planning method and a graph convolutional network (GCN)-Transformer Link Prediction retrieval model, which enhances the accuracy and reliability of the generated simulations. The AutoFEA system proceeds with key steps: dataset preparation, step-by-step planning, GCN-Transformer Link Prediction retrieval, LLM-driven code generation, and simulation using CalculiX. In this workflow, the GCN-Transformer model predicts and retrieves relevant example codes based on relationships between different steps in the FEA process, guiding the LLM in generating accurate simulation codes. We validate AutoFEA using a specialized dataset of 512 meticulously prepared FEA projects, which provides a robust foundation for training and evaluation. Our results demonstrate that AutoFEA significantly reduces AI hallucinations by grounding LLM outputs in physically accurate simulation data, thereby improving the success rate and accuracy of FEA simulations and paving the way for future advancements in AI-assisted engineering tasks.

Extreme events frequently occur in real-world time series and often carry significant practical implications. In domains such as climate and healthcare, these events, such as floods, heatwaves, or acute medical episodes, can lead to serious consequences. Accurate forecasting of such events is therefore of substantial importance. Most existing time series forecasting models are optimized for overall performance within the prediction window, but often struggle to accurately predict extreme events, such as high temperatures or heart rate spikes. The main challenges are data imbalance and the neglect of valuable information contained in intermediate events that precede extreme events. In this paper, we propose xTime, a novel framework for extreme event forecasting in time series. xTime leverages knowledge distillation to transfer information from models trained on lower-rarity events, thereby improving prediction performance on rarer ones. In addition, we introduce a MoE mechanism that dynamically selects and fuses outputs from expert models across different rarity levels, which further improves the forecasting performance for extreme events. Experiments on multiple datasets show that xTime achieves consistent improvements, with forecasting accuracy on extreme events improving from 3% to 78%.

Graphs are pervasive in the real-world, such as social network analysis, bioinformatics, and knowledge graphs. Graph neural networks (GNNs) have great ability in node classification but still face challenges in scenarios with few labeled nodes, despite the frequent occurrence of such tasks in real-world applications with limited labeled data. To address this challenge, various approaches have been proposed, including graph meta-learning, transfer learning, and methods based on Large Language Models (LLMs). However, traditional meta-learning and transfer learning methods often require prior knowledge from base classes or fail to exploit the potential advantages of unlabeled nodes. Meanwhile, LLM-based methods may overlook the zero-shot capabilities of LLMs and rely heavily on the quality of generated contexts. In this paper, we propose a novel approach that integrates LLMs and GNNs, leveraging the zero-shot inference capabilities of LLMs and employing a Graph-LLM-based active learning paradigm to enhance GNNs’ performance. Extensive experiments demonstrate the effectiveness of our model in improving node classification accuracy with considerably limited labeled data, surpassing state-of-the-art baselines by significant margins.

As fraudulent activities have shot up manifolds, fraud detection has emerged as a pivotal process in different fields (e.g., e-commerce, online reviews, and social networks). Since interactions among entities provide valuable insights into fraudulent activities, such behaviors can be naturally represented as graph structures, where graph neural networks (GNNs) have been developed as prominent models to boost the efficacy of fraud detection. In graph-based fraud detection, handling imbalanced datasets poses a significant challenge, as the minority class often gets overshadowed, diminishing the performance of conventional GNNs. While oversampling has recently been adapted for imbalanced graphs, it contends with issues such as graph heterophily and noisy edge synthesis. To address these limitations, this paper introduces DOS-GNN, incorporating Dual-feature aggregation with Over-Sampling to advance GNNs for class-imbalanced fraud detection on graphs. This model exploits feature separation and dual-feature aggregation to mitigate the impact of heterophily and acquire refined node embeddings that facilitate fraud oversampling to balance class distribution without the need for edge synthesis. Extensive experiments on four large and real-world fraud datasets demonstrate that DOS-GNN can significantly improve fraud detection performance on graphs with different imbalance ratios and homophily ratios, outperforming state-of-the-art GNN models.

Malware developers often employ code obfuscation techniques to conceal their malicious functionality, making it challenging to detect and analyze such software. While various de-obfuscation techniques exist, the majority of them require prior knowledge of the obfuscation tools and techniques in use. Identifying the specific obfuscation tools or algorithms applied to the obfuscated code is thus of vital importance, which, however, typically demands in-depth expert knowledge and substantial efforts. Therefore, this paper presents DeObA, a deep learning (DL) driven approach for the precise and efficient detection of obfuscation algorithms on the fine-grained function-level code snippets. To comprehensively capture unique patterns or features of different obfuscation algorithms from code, DeObA works on multiple distinct code views, encompassing token sequences, abstract syntax trees (AST) and program dependency graphs (PDG), which will reflect the code’s lexical morphology, syntactic and structural aspects. After individually collecting obfuscation-indicative features with well-matched DL encoder from each code view, a self-attention-based fusion strategy is performed on these features to produce an integrated, dense, yet feature-rich vector. This vector is then fed into a softmax classification layer for prediction. Due to the lack of a moderately sized dataset, a large obfuscation corpus is curated with 7 different obfuscation tools and a total of 12 obfuscation algorithms on 39,070 C/C[Formula: see text] functions. The experimental evaluations conducted on the dataset exhibit a distinguished detection performance of DeObA, which achieve accuracy rates of 99.90% and 99.19% on the obfuscation tool detection and obfuscation algorithm detection tasks, respectively. The ablation study also confirms the active role of considering multiple distinct code views and the effectiveness of the designed self-attention-based fusion strategy.

Social media has drastically reshaped the world that allows billions of people to engage in such interactive environments to conveniently create and share content with the public. Among them, text data (e.g., tweets, blogs) maintains the basic yet important social activities and generates a rich source of user-oriented information. While those explicit sensitive user data like credentials have been significantly protected by all means, personal private attribute (e.g., age, gender, location) disclosure due to inference attacks is somehow challenging to avoid, especially when powerful natural language processing (NLP) techniques have been effectively deployed to automate attribute inferences from implicit text data. This puts users’ attribute privacy at risk. To address this challenge, in this article, we leverage the inherent vulnerability of machine learning to adversarial attacks, and design a novel text-space Adv ersarial attack for S ocial G ood, called Adv4SG . In other words, we cast the problem of protecting personal attribute privacy as an adversarial attack formulation problem over the social media text data to defend against NLP-based attribute inference attacks. More specifically, Adv4SG proceeds with a sequence of word perturbations under given constraints such that the probed attribute cannot be identified correctly. Different from the prior works, we advance Adv4SG by considering social media property, and introducing cost-effective mechanisms to expedite attribute obfuscation over text data under the black-box setting. Extensive experiments on real-world social media datasets have demonstrated that our method can effectively degrade the inference accuracy with less computational cost over different attribute settings, which substantially helps mitigate the impacts of inference attacks and thus achieve high performance in user attribute privacy protection.

As malicious bots reside in a network to disrupt network stability, graph neural networks (GNNs) have emerged as one of the most popular bot detection methods. However, in most cases these graphs are significantly class-imbalanced. To address this issue, graph oversampling has recently been proposed to synthesize nodes and edges, which still suffers from graph heterophily, leading to suboptimal performance. In this paper, we propose HOVER, which implements Homophilic Oversampling Via Edge Removal for bot detection on graphs. Instead of oversampling nodes and edges within initial graph structure, HOVER designs a simple edge removal method with heuristic criteria to mitigate heterophily and learn distinguishable node embeddings, which are then used to oversample minority bots to generate a balanced class distribution without edge synthesis. Experiments on TON IoT networks demonstrate the state-of-the-art performance of HOVER on bot detection with high graph heterophily and extreme class imbalance.

LIME has emerged as one of the most commonly referenced tools in explainable AI (XAI) frameworks that is integrated into critical machine learning applications-e.g., healthcare and finance.However, its stability remains little explored, especially in the context of text data, due to the unique text-space constraints.To address these challenges, in this paper, we first evaluate the inherent instability of LIME on text data to establish a baseline, and then propose a novel algorithm XAIFOOLER to perturb text inputs and manipulate explanations that casts investigation on the stability of LIME as a text perturbation optimization problem.XAIFOOLER conforms to the constraints to preserve text semantics and original prediction with small perturbations, and introduces Rankbiased Overlap (RBO) as a key part to guide the optimization of XAIFOOLER that satisfies all the requirements for explanation similarity measure.Extensive experiments on real-world text datasets demonstrate that XAIFOOLER significantly outperforms all baselines by large margins in its ability to manipulate LIME's explanations with high semantic preservability.The code is available at https://github. com/cburgerOlemiss/XAIFooler

Graphs have emerged as one of the most important and powerful data structures to perform content analysis in many fields. In this line of work, node classification is a classic task, which is generally performed using graph neural networks (GNNs). Unfortunately, regular GNNs cannot be well generalized into the real-world application scenario when the labeled nodes are few. To address this challenge, we propose a novel few-shot node classification model that leverages pseudo-labeling with graph active learning. We first provide a theoretical analysis to argue that extra unlabeled data benefit few-shot classification. Inspired by this, our model proceeds by performing multi-level data augmentation with consistency and contrastive regularizations for better semi-supervised pseudo-labeling, and further devising graph active learning to facilitate pseudo-label selection and improve model effectiveness. Extensive experiments on four public citation networks have demonstrated that our model can effectively improve node classification accuracy with considerably few labeled data, which significantly outperforms all state-of-the-art baselines by large margins.

In this paper, we consider how the ability to learn Code Execution Tasks affects a model’s accuracy on software vulnerability detection (SVD) benchmark datasets. We initially find that models can achieve near state-of-the-art accuracy on SVD benchmarks regardless of their ability to learn Code Execution Tasks. However, these models fail to generalize well across SVD benchmarks. The results indicate a bias in the datasets that allows models to predict non-SVD signals. Under the theory that different collection methods will reduce biases, we investigate combining the SVD datasets. When trained on combined datasets, SVD accuracy is reduced but correlation with Code Execution Task accuracy improves. Our contributions are (1) using a reversed curriculum learning to evaluate model capabilities, (2) demonstrating the criticality of code execution understanding to machine learning–assisted software vulnerability detection, (3) evidence that improved diversity of SVD datasets will lead to improved accuracy and generalizability, (4) and benchmarks of recent models across multiple SVD datasets.

Fuzzing is a widely-used testing technique to assure software robustness. However, automatic generation of high-quality test suites is challenging, especially for software that takes in highly-structured inputs, such as the compilers. Compiler fuzzing remains difficult as generating tons of syntactically and semantically valid programs is not trivial. Most previous methods either depend on human-crafted grammars or heuristics to learn partial language patterns. They both suffer from the completeness issue that is a classic puzzle in software testing. To mitigate the problem, we propose a knowledge-guided reinforcement learning-based approach to generating valid programs for compiler fuzzing. We first design a naive learning model which evolves with the sequential mutation rewards provided by a target compiler we test. By iterating the training cycle, the model learns to generate valid programs that can improve the testing efficacy as well. We implement the proposed method into a tool called ALPHAPROG. We analyze the framework with four different reward functions and our study reveal the effectiveness of ALPHAPROG for compiler testing. We also reported two important bugs for a compiler production that were confirmed and addressed by the project owner, which further demonstrates ALPHAPROG's applied value in practice.

Modern software developers tend to engage in social coding platforms to reuse code snippets to expedite the development process, while the codes on such platforms are often suffering from comments being mismatched, missing or outdated. This puts the code search and comprehension in difficulty, and increases the burden of maintenance for software building upon these codes. As summarizing code is beneficial yet it is very expensive for manual operation, in this paper, we elaborate an automatic and effective code summarization paradigm to address this laborious challenge. We represent a given code snippet as an abstract syntax tree (AST), and generate a set of compositional root-to-leaf paths to make the AST accessible regarding code context and structure in a less complex yet expressive way. Accordingly, we design a tree-based transformer model, called TreeXFMR, on these paths to summarize source code in a hierarchical attention operation. This yields two advantages on code representation learning: (1) attention mechanisms at token-and path-level attend the semantics and interactions of source code from different aspects; (2) bi-level positional encodings introduced reveal the intra- and inter-path structure of AST and improve the unambiguity of the representations. During decoding, TreeXFMR attends such learned representations to produce each output of natural language word. We further pre-train the transformer to achieve faster and better training convergence results. Extensive experiments on the code collection from GitHub demonstrate the effectiveness of TreeXFMR, which significantly outperforms state-of-the-art baselines.

To mitigate evolving malware attacks, machine learning models have been successfully deployed to detect malware. However, these models are often challenged by data scarcity, design efforts and constrained resources. Inspired by the adversarial vulnerability of machine learning, in this paper, we design a novel model Adv4Mal to adversarially reprogram an ImageNet classification neural network for malware detection in both white-box and black-box settings. As such, a small or moderate amount of data are sufficient to train a promising malware detection model, the varying software features can be uniformly processed without extra efforts, and the majority of computation can be wisely shared and reused to save the resources. This, to the best of our knowledge, has not yet been explored. Specifically, Adv4Mal proceeds by embedding software features into a host image to construct new data, and learning a universal perturbation to be added to all inputs in an imperceptible manner, such that the outputs of the pretrained model can be accordingly mapped to the final detection decisions for all software. We evaluate Adv4Mal on three software datasets. The experimental results demonstrate that Adv4Mal can successfully exploit ImageNet model's learning capability and limited data to achieve high performance in malware detection, and also yield significant advantages of model flexibility to different features, and cost efficiency in computing resources.

Smart contracts are programs that run on a blockchain, where Ethereum is one of the most popular ones supporting them.Due to the fact that they are immutable, it is essential to design smart contracts bug-free before they are deployed.However, various defects have been found in the deployed smart contracts, causing huge economic losses and lowing people's trust.Writing secure smart contracts is far from trivial, where developers tend to engage in reliable resources or social coding platforms to reuse code.This leads to a large number of similar contracts with potential security risks.Therefore, detecting similarity of smart contracts helps to avoid vulnerabilities, identify threats, and improve the security of Ethereum.In this paper, we design a learning-effective and costefficient model, called SmartSD, for Ethereum smart contract similarity detection.Different from the current research efforts, SmartSD is performed on a bytecode level and leverages deep neural networks to learn the latent representations from the opcode sequences for smart contract bytecodes, where the representation learning and similarity measurement are supervised via siamese neural networks.The experimental evaluations demonstrate that SmartSD outperforms EClone's 93.27% accuracy, achieving 98.37% high detection accuracy and 0.9850 F1-score, which is computationally tractable and effectively mitigates the interference caused by compilers.

The popularization of social media generates a large amount of user-oriented data, where text data especially attracts researchers and speculators to infer user attributes (e.g., age, gender) for fulfilling their intents. Generally, this line of work casts attribute inference as a text classification problem, and starts to leverage graph neural networks for higher-level text representations. However, these text graphs are constructed on words, suffering from high memory consumption and ineffectiveness on few labeled texts. To address this challenge, we design a text-graph-based few-shot learning model for social media attribute inferences. Our model builds a text graph with texts as nodes and edges learned from current text representations via manifold learning and message passing. To further use unlabeled texts to improve few-shot performance, a knowledge distillation is devised to optimize the problem. This offers a trade-off between expressiveness and complexity. Experiments on social media datasets demonstrate the state-of-the-art performance of our model on attribute inferences with considerably fewer labeled texts.

Machine learning models have been widely used for fraud detection, while developing and maintaining these models often suffers from significant limitations in terms of training data scarcity and constrained resources. To address these issues, in this paper, we leverage machine learning vulnerability to adversarial attacks, and design a novel model AdvRFD that Adversarially Reprograms an ImageNet classification neural network for Fraud Detection task. AdvRFD first embeds transaction features into a host image to construct new ImageNet data, and then learns a universal perturbation to be added to all inputs, such that the outputs of the pretrained model can be accordingly mapped to the final detection decisions for all transactions. Extensive experiments on two transaction datasets made over Ethereum and credit cards have demonstrated that AdvRFD is effective to detect fraud using limited data and resources.

Few-shot Learning, a recent research hotspot in computer vision, attempts to distinguish unseen classes with a few labeled samples. Although significant progress has been made in this field, difficulties still exist in promoting classification accuracy. To tackle this issue, we proposed an effective and novel approach called compare network with task-dependent embeddings for few-shot learning, which includes three modules: embedding module, self-attention module and compare module. With the features extracted by the embedding module, the self-attention module, a use of attention mechanisms as component, can fully capture the internal relevance of features in the context of a certain task so as to learn more discriminative and adaptive task-dependent embeddings by taking a view of the entire task. And the compare module, the direct use of attention mechanisms as metrics, can efficiently evaluate the similarity score between query and discriminative task-dependent embeddings. The effectiveness of our architecture has been verified on two standard classification benchmark, namely the miniImageNet and Omniglot. Experimental results show that our method achieved very competitive results when compared to the state-of-the-art method in terms of few-shot tasks and realized consistent improvements over baselines.

The vulnerability of deep neural networks to adversarial attacks has posed significant threats to real-world applications, especially security-critical ones. Given a well-trained model, slight modifications to the input samples can cause drastic changes in the predictions of the model. Many methods have been proposed to mitigate the issue. However, the majority of these defenses have proven to fail to resist all the adversarial attacks. This is mainly because the knowledge advantage of the attacker can help to either easily customize the information of the target model or create a surrogate model as a substitute to successfully construct the corresponding adversarial examples. In this paper, we propose a new defense mechanism that creates a knowledge gap between attackers and defenders by imposing a designed watermarking system into standard deep neural networks. The embedded watermark is data-independent and non-reproducible to an attacker, which improves randomization and security of the defense model without compromising performance on clean data, and thus yields knowledge disadvantage to prevent an attacker from crafting effective adversarial examples targeting the defensive model. We evaluate the performance of our watermarking defense using a wide range of watermarking algorithms against four state-of-the-art attacks on different datasets, and the experimental results validate its effectiveness.

Machine learning, especially deep learning, has emerged as one of the most powerful tools for attribute inference attacks over social media, which poses serious threats to users' privacy and security. In this paper, we explore a novel perspective of protecting data privacy in social media, where we take advantage of the vulnerability of machine learning, and introduce adversarial attacks to forge latent feature representations and mislead attribute inference attacks. Considering that text data in social media shares the most significant privacy of users, we investigate how text-space adversarial attacks can be elaborated to obfuscate users' attributes, and accordingly present a text-space adversarial attack as defense, or AaaD for short. Specifically, we advance AaaD by constructing semantically and visually similar word candidates to perturb, and leveraging word importance scores as selection probabilities to upgrade a population-based optimization to expedite adversarial text generation. We evaluate the performance of AaaD on two social media data sets, while the experimental results validate its effectiveness against inference attacks. Our work yields great value and unveils a new insight on the applicability of adversarial attacks for attribute obfuscation and privacy protection.

Software dynamic birthmark techniques construct birthmarks using the captured execution traces from running the programs, which serve as one of the most promising methods for obfuscation-resilient software plagiarism detection. However, due to the perturbation caused by non-deterministic thread scheduling in multi-threaded programs, such dynamic approaches optimized for sequential programs may suffer from the randomness in multi-threaded program plagiarism detection. In this paper, we propose a new dynamic thread-aware birthmark FPBirth to facilitate multi-threaded program plagiarism detection. We first explore dynamic monitoring to capture multiple execution traces with respect to system calls for each multi-threaded program under a specified input, and then leverage the Apriori algorithm to mine frequent patterns to formulate our dynamic birthmark, which can not only depict the program’s behavioral semantics, but also resist the changes and perturbations over execution traces caused by the thread scheduling in multi-threaded programs. Using FPBirth, we design a multi-threaded program plagiarism detection system. The experimental results based on a public software plagiarism sample set demonstrate that the developed system integrating our proposed birthmark FPBirth copes better with multi-threaded plagiarism detection than alternative approaches. Compared against the dynamic birthmark System Call Short Sequence Birthmark (SCSSB), FPBirth achieves 12.4%, 4.1% and 7.9% performance improvements with respect to union of resilience and credibility (URC), F-Measure and matthews correlation coefficient (MCC) metric, respectively.

Despite the apparent benefits of modern social coding paradigm such as Stack Overflow, its potential security risks have been largely overlooked (e.g., insecure codes could be easily embedded and distributed). To address this imminent issue, in this paper, we bring a significant insight to leverage both social coding properties and code content for automatic detection of insecure code snippets in Stack Overflow. To determine if the given code snippets are insecure, we not only analyze the code content, but also utilize various kinds of relations among users, badges, questions, answers and code snippets in Stack Overflow. To model the rich semantic relationships, we first introduce a structured heterogeneous information network (HIN) for representation and then use meta-path based approach to incorporate higher-level semantics to build up relatedness over code snippets. Later, we propose a novel hierarchical attention-based sequence learning model named CodeHin2Vec to seamlessly integrate node (i.e., code snippet) content with HIN-based relations for representation learning. After that, a classifier is built for insecure code snippet detection. Integrating our proposed method, an intelligent system named iTrustSO is accordingly developed to address the code security issues in modern software coding platforms. Comprehensive experiments on the data collections from Stack Overflow are conducted to validate the effectiveness of our developed system iTrustSO by comparisons with alternative methods.

As modern social coding platforms such as GitHub and Stack Overflow become increasingly popular, their potential security risks increase as well (e.g., risky or malicious codes could be easily embedded and distributed). To enhance the social coding security, in this paper, we propose to automate cross-platform user identification between GitHub and Stack Overflow to combat the attackers who attempt to poison the modern software programming ecosystem. To solve this problem, an important insight brought by this work is to leverage social coding properties in addition to user attributes for cross-platform user identification. To depict users in GitHub and Stack Overflow (attached with attributed information), projects, questions and answers as well as the rich semantic relations among them, we first introduce an attributed heterogeneous information network (AHIN) for modeling. Then, we propose a novel AHIN representation learning model AHIN2Vec to efficiently learn node (i.e., user) representations in AHIN for cross-platform user identification. Comprehensive experiments on the data collections from GitHub and Stack Overflow are conducted to validate the effectiveness of our developed system iDev integrating our proposed method in cross-platform user identification by comparisons with other baselines.

The increasingly sophisticated Android malware calls for new defensive techniques that are capable of protecting mobile users against novel threats. In this paper, we first extract the runtime Application Programming Interface (API) call sequences from Android apps, and then analyze higher-level semantic relations within the ecosystem to comprehensively characterize the apps. To model different types of entities (i.e., app, API, device, signature, affiliation) and rich relations among them, we present a structured heterogeneous graph (HG) for modeling. To efficiently classify nodes (e.g., apps) in the constructed HG, we propose the HG-Learning method to first obtain in-sample node embeddings and then learn representations of out-of-sample nodes without rerunning/adjusting HG embeddings at the first attempt. We later design a deep neural network classifier taking the learned HG representations as inputs for real-time Android malware detection. Comprehensive experiments on large-scale and real sample collections from Tencent Security Lab are performed to compare various baselines. Promising results demonstrate that our developed system AiDroid which integrates our proposed method outperforms others in real-time Android malware detection.

As the popularity of modern social coding paradigm such as Stack Overflow grows, its potential security risks increase as well (e.g., insecure codes could be easily embedded and distributed). To address this largely overlooked issue, in this paper, we bring an important new insight to exploit social coding properties in addition to code content for automatic detection of insecure code snippets in Stack Overflow. To determine if the given code snippets are insecure, we not only analyze the code content, but also utilize various kinds of relations among users, badges, questions, answers, code snippets and keywords in Stack Overflow. To model the rich semantic relationships, we first introduce a structured heterogeneous information network (HIN) for representation and then use meta-path based approach to incorporate higher-level semantics to build up relatedness over code snippets. Later, we propose a novel network embedding model named snippet2vec for representation learning in HIN where both the HIN structures and semantics are maximally preserved. After that, a multi-view fusion classifier is constructed for insecure code snippet detection. To the best of our knowledge, this is the first work utilizing both code content and social coding properties to address the code security issues in modern software coding platforms. Comprehensive experiments on the data collections from Stack Overflow are conducted to validate the effectiveness of the developed system ICSD which integrates our proposed method in insecure code snippet detection by comparisons with alternative approaches.

To combat the evolving Android malware attacks, systems using machine learning techniques have been successfully deployed for Android malware detection. In these systems, based on different feature representations, various kinds of classifiers are constructed to detect Android malware. Unfortunately, as classifiers become more widely deployed, the incentive for defeating them increases. In this paper, we first extract a set of features from the Android applications (apps) and represent them as binary feature vectors; with these inputs, we then explore the security of a generic learning-based classifier for Android malware detection in the presence of adversaries. To harden the evasion, we first present count featurization to transform the binary feature space into continuous probabilities encoding the distribution in each class (either benign or malicious). To improve the system security while not compromising the detection accuracy, we further introduce softmax function with adversarial parameter to find the best trade-off between security and accuracy for the classifier. Accordingly, we develop a system named DroidEye which integrates our proposed method for Android malware detection. Comprehensive experiments on the real sample collection from Comodo Cloud Security Center are conducted to validate the effectiveness of DroidEye against adversarial Android malware attacks. Our proposed secure-learning paradigm is also applicable for other detection tasks, such as spammer detection in social media.

With smart phones being indispensable in people's everyday life, Android malware has posed serious threats to their security, making its detection of utmost concern. To protect legitimate users from the evolving Android malware attacks, machine learning-based systems have been successfully deployed and offer unparalleled flexibility in automatic Android malware detection. In these systems, based on different feature representations, various kinds of classifiers are constructed to detect Android malware. Unfortunately, as classifiers become more widely deployed, the incentive for defeating them increases. In this paper, we explore the security of machine learning in Android malware detection on the basis of a learning-based classifier with the input of a set of features extracted from the Android applications (apps). We consider different importances of the features associated with their contributions to the classification problem as well as their manipulation costs, and present a novel feature selection method (named SecCLS) to make the classifier harder to be evaded. To improve the system security while not compromising the detection accuracy, we further propose an ensemble learning approach (named SecENS) by aggregating the individual classifiers that are constructed using our proposed feature selection method SecCLS. Accordingly, we develop a system called SecureDroid which integrates our proposed methods (i.e., SecCLS and SecENS) to enhance security of machine learning-based Android malware detection. Comprehensive experiments on the real sample collections from Comodo Cloud Security Center are conducted to validate the effectiveness of SecureDroid against adversarial Android malware attacks by comparisons with other alternative defense methods. Our proposed secure-learning paradigm can also be readily applied to other malware detection tasks.

Because of the explosive growth of Android malware and due to the severity of its damages, the detection of Android malware has become an increasing important topic in cybersecurity. Currently, the major defense against Android malware is commercial mobile security products which mainly use signature-based method for detection. However, attackers can easily devise methods, such as obfuscation and repackaging, to evade the detection, which calls for new defensive techniques that are harder to evade. In this paper, resting on the analysis of Application Programming Interface (API) calls extracted from the smali files, we further categorize the API calls which belong to the some method in the smali code into a block. Based on the generated API call blocks, we then explore deep neural networks (i.e., Deep Belief Network (DBN) and Stacked AutoEncoders (SAEs)) for newly unknown Android malware detection. Using a real sample collection from Comodo Cloud Security Center, a comprehensive experimental study is performed to compare various malware detection approaches. The experimental results demonstrate that (1) our proposed feature extraction method (i.e., using API call blocks) outperforms using API calls directly in Android malware detection; (2) DBN works better than SAEs in this application; and (3) the detection performance of deep neural networks is better than shallow learning architectures.