Yves Le Traon

Qiang Hu, Yuejun Guo, Maxime Cordy et al.

Deep Neural Networks (DNNs) have gained considerable attention in the past decades due to their astounding performance in different applications, such as natural language modeling, self-driving assistance, and source code understanding. With rapid exploration, more and more complex DNN architectures have been proposed along with huge pre-trained model parameters. The common way to use such DNN models in user-friendly devices (e.g., mobile phones) is to perform model compression before deployment. However, recent research has demonstrated that model compression, e.g., model quantization, yields accuracy degradation as well as outputs disagreements when tested on unseen data. Since the unseen data always include distribution shifts and often appear in the wild, the quality and reliability of quantized models are not ensured. In this paper, we conduct a comprehensive study to characterize and help users understand the behaviors of quantized models. Our study considers 4 datasets spanning from image to text, 8 DNN architectures including feed-forward neural networks and recurrent neural networks, and 42 shifted sets with both synthetic and natural distribution shifts. The results reveal that 1) data with distribution shifts happen more disagreements than without. 2) Quantization-aware training can produce more stable models than standard, adversarial, and Mixup training. 3) Disagreements often have closer top-1 and top-2 output probabilities, and $Margin$ is a better indicator than the other uncertainty metrics to distinguish disagreements. 4) Retraining with disagreements has limited efficiency in removing disagreements. We opensource our code and models as a new benchmark for further studying the quantized models.

Qiang Hu, Yuejun Guo, Maxime Cordy et al.

Applying deep learning to science is a new trend in recent years which leads DL engineering to become an important problem. Although training data preparation, model architecture design, and model training are the normal processes to build DL models, all of them are complex and costly. Therefore, reusing the open-sourced pre-trained model is a practical way to bypass this hurdle for developers. Given a specific task, developers can collect massive pre-trained deep neural networks from public sources for re-using. However, testing the performance (e.g., accuracy and robustness) of multiple DNNs and recommending which model should be used is challenging regarding the scarcity of labeled data and the demand for domain expertise. In this paper, we propose a labeling-free (LaF) model selection approach to overcome the limitations of labeling efforts for automated model reusing. The main idea is to statistically learn a Bayesian model to infer the models' specialty only based on predicted labels. We evaluate LaF using 9 benchmark datasets including image, text, and source code, and 165 DNNs, considering both the accuracy and robustness of models. The experimental results demonstrate that LaF outperforms the baseline methods by up to 0.74 and 0.53 on Spearman's correlation and Kendall's $τ$, respectively.

Yuejun Guo, Qiang Hu, Qiang Tang et al.

Vulnerability detection is crucial to protect software security. Nowadays, deep learning (DL) is the most promising technique to automate this detection task, leveraging its superior ability to extract patterns and representations within extensive code volumes. Despite its promise, DL-based vulnerability detection remains in its early stages, with model performance exhibiting variability across datasets. Drawing insights from other well-explored application areas like computer vision, we conjecture that the imbalance issue (the number of vulnerable code is extremely small) is at the core of the phenomenon. To validate this, we conduct a comprehensive empirical study involving nine open-source datasets and two state-of-the-art DL models. The results confirm our conjecture. We also obtain insightful findings on how existing imbalance solutions perform in vulnerability detection. It turns out that these solutions perform differently as well across datasets and evaluation metrics. Specifically: 1) Focal loss is more suitable to improve the precision, 2) mean false error and class-balanced loss encourages the recall, and 3) random over-sampling facilitates the F1-measure. However, none of them excels across all metrics. To delve deeper, we explore external influences on these solutions and offer insights for developing new solutions.

Martin Gubri, Maxime Cordy, Mike Papadakis et al.

We propose transferability from Large Geometric Vicinity (LGV), a new technique to increase the transferability of black-box adversarial attacks. LGV starts from a pretrained surrogate model and collects multiple weight sets from a few additional training epochs with a constant and high learning rate. LGV exploits two geometric properties that we relate to transferability. First, models that belong to a wider weight optimum are better surrogates. Second, we identify a subspace able to generate an effective surrogate ensemble among this wider optimum. Through extensive experiments, we show that LGV alone outperforms all (combinations of) four established test-time transformations by 1.8 to 59.9 percentage points. Our findings shed new light on the importance of the geometry of the weight space to explain the transferability of adversarial examples.

Qiang Hu, Yuejun Guo, Xiaofei Xie et al.

Deep learning (DL) plays a more and more important role in our daily life due to its competitive performance in industrial application domains. As the core of DL-enabled systems, deep neural networks (DNNs) need to be carefully evaluated to ensure the produced models match the expected requirements. In practice, the \emph{de facto standard} to assess the quality of DNNs in the industry is to check their performance (accuracy) on a collected set of labeled test data. However, preparing such labeled data is often not easy partly because of the huge labeling effort, i.e., data labeling is labor-intensive, especially with the massive new incoming unlabeled data every day. Recent studies show that test selection for DNN is a promising direction that tackles this issue by selecting minimal representative data to label and using these data to assess the model. However, it still requires human effort and cannot be automatic. In this paper, we propose a novel technique, named \textit{Aries}, that can estimate the performance of DNNs on new unlabeled data using only the information obtained from the original test data. The key insight behind our technique is that the model should have similar prediction accuracy on the data which have similar distances to the decision boundary. We performed a large-scale evaluation of our technique on two famous datasets, CIFAR-10 and Tiny-ImageNet, four widely studied DNN models including ResNet101 and DenseNet121, and 13 types of data transformation methods. Results show that the estimated accuracy by \textit{Aries} is only 0.03\% -- 2.60\% off the true accuracy. Besides, \textit{Aries} also outperforms the state-of-the-art labeling-free methods in 50 out of 52 cases and selection-labeling-based methods in 96 out of 128 cases.

Salah Ghamizi, Jingfeng Zhang, Maxime Cordy et al.

While leveraging additional training data is well established to improve adversarial robustness, it incurs the unavoidable cost of data collection and the heavy computation to train models. To mitigate the costs, we propose Guided Adversarial Training (GAT), a novel adversarial training technique that exploits auxiliary tasks under a limited set of training data. Our approach extends single-task models into multi-task models during the min-max optimization of adversarial training, and drives the loss optimization with a regularization of the gradient curvature across multiple tasks. GAT leverages two types of auxiliary tasks: self-supervised tasks, where the labels are generated automatically, and domain-knowledge tasks, where human experts provide additional labels. Experimentally, GAT increases the robust AUC of CheXpert medical imaging dataset from 50% to 83% and On CIFAR-10, GAT outperforms eight state-of-the-art adversarial training and achieves 56.21% robust accuracy with Resnet-50. Overall, we demonstrate that guided multi-task learning is an actionable and promising avenue to push further the boundaries of model robustness.

Zeming Dong, Qiang Hu, Yuejun Guo et al.

Inspired by the great success of Deep Neural Networks (DNNs) in natural language processing (NLP), DNNs have been increasingly applied in source code analysis and attracted significant attention from the software engineering community. Due to its data-driven nature, a DNN model requires massive and high-quality labeled training data to achieve expert-level performance. Collecting such data is often not hard, but the labeling process is notoriously laborious. The task of DNN-based code analysis even worsens the situation because source code labeling also demands sophisticated expertise. Data augmentation has been a popular approach to supplement training data in domains such as computer vision and NLP. However, existing data augmentation approaches in code analysis adopt simple methods, such as data transformation and adversarial example generation, thus bringing limited performance superiority. In this paper, we propose a data augmentation approach MIXCODE that aims to effectively supplement valid training data, inspired by the recent advance named Mixup in computer vision. Specifically, we first utilize multiple code refactoring methods to generate transformed code that holds consistent labels with the original data. Then, we adapt the Mixup technique to mix the original code with the transformed code to augment the training data. We evaluate MIXCODE on two programming languages (Java and Python), two code tasks (problem classification and bug detection), four benchmark datasets (JAVA250, Python800, CodRep1, and Refactory), and seven model architectures (including two pre-trained models CodeBERT and GraphCodeBERT). Experimental results demonstrate that MIXCODE outperforms the baseline data augmentation approach by up to 6.24% in accuracy and 26.06% in robustness.

Qiang Hu, Yuejun Guo, Xiaofei Xie et al.

Distribution shift has been a longstanding challenge for the reliable deployment of deep learning (DL) models due to unexpected accuracy degradation. Although DL has been becoming a driving force for large-scale source code analysis in the big code era, limited progress has been made on distribution shift analysis and benchmarking for source code tasks. To fill this gap, this paper initiates to propose CodeS, a distribution shift benchmark dataset, for source code learning. Specifically, CodeS supports two programming languages (Java and Python) and five shift types (task, programmer, time-stamp, token, and concrete syntax tree). Extensive experiments based on CodeS reveal that 1) out-of-distribution detectors from other domains (e.g., computer vision) do not generalize to source code, 2) all code classification models suffer from distribution shifts, 3) representation-based shifts have a higher impact on the model than others, and 4) pre-trained bimodal models are relatively more resistant to distribution shifts.

Zeming Dong, Qiang Hu, Yuejun Guo et al.

Recent studies have demonstrated remarkable advancements in source code learning, which applies deep neural networks (DNNs) to tackle various software engineering tasks. Similar to other DNN-based domains, source code learning also requires massive high-quality training data to achieve the success of these applications. Data augmentation, a technique used to produce additional training data, is widely adopted in other domains (e.g. computer vision). However, the existing practice of data augmentation in source code learning is limited to simple syntax-preserved methods, such as code refactoring. In this paper, considering that source code can also be represented as text data, we take an early step to investigate the effectiveness of data augmentation methods originally designed for natural language texts in the context of source code learning. To this end, we focus on code classification tasks and conduct a comprehensive empirical study across four critical code problems and four DNN architectures to assess the effectiveness of 25 data augmentation methods. Our results reveal specific data augmentation methods that yield more accurate and robust models for source code learning. Additionally, we discover that the data augmentation methods remain beneficial even when they slightly break source code syntax.

Ezekiel Soremekun, Mike Papadakis, Maxime Cordy et al.

In the last decade, researchers have studied fairness as a software property. In particular, how to engineer fair software systems? This includes specifying, designing, and validating fairness properties. However, the landscape of works addressing bias as a software engineering concern is unclear, i.e., techniques and studies that analyze the fairness properties of learning-based software. In this work, we provide a clear view of the state-of-the-art in software fairness analysis. To this end, we collect, categorize and conduct an in-depth analysis of 164 publications investigating the fairness of learning-based software systems. Specifically, we study the evaluated fairness measure, the studied tasks, the type of fairness analysis, the main idea of the proposed approaches, and the access level (e.g., black, white, or grey box). Our findings include the following: (1) Fairness concerns (such as fairness specification and requirements engineering) are under-studied; (2) Fairness measures such as conditional, sequential, and intersectional fairness are under-explored; (3) Unstructured datasets (e.g., audio, image, and text) are barely studied for fairness analysis; and (4) Software fairness analysis techniques hardly employ white-box, in-processing machine learning (ML) analysis methods. In summary, we observed several open challenges including the need to study intersectional/sequential bias, policy-based bias handling, and human-in-the-loop, socio-technical bias mitigation.

Qiang Hu, Yuejun Guo, Xiaofei Xie et al.

Testing deep learning-based systems is crucial but challenging due to the required time and labor for labeling collected raw data. To alleviate the labeling effort, multiple test selection methods have been proposed where only a subset of test data needs to be labeled while satisfying testing requirements. However, we observe that such methods with reported promising results are only evaluated under simple scenarios, e.g., testing on original test data. This brings a question to us: are they always reliable? In this paper, we explore when and to what extent test selection methods fail for testing. Specifically, first, we identify potential pitfalls of 11 selection methods from top-tier venues based on their construction. Second, we conduct a study on five datasets with two model architectures per dataset to empirically confirm the existence of these pitfalls. Furthermore, we demonstrate how pitfalls can break the reliability of these methods. Concretely, methods for fault detection suffer from test data that are: 1) correctly classified but uncertain, or 2) misclassified but confident. Remarkably, the test relative coverage achieved by such methods drops by up to 86.85%. On the other hand, methods for performance estimation are sensitive to the choice of intermediate-layer output. The effectiveness of such methods can be even worse than random selection when using an inappropriate layer.

Zeming Dong, Qiang Hu, Zhenya Zhang et al.

Graph neural network (GNN)-based graph learning has been popular in natural language and programming language processing, particularly in text and source code classification. Typically, GNNs are constructed by incorporating alternating layers which learn transformations of graph node features, along with graph pooling layers that use graph pooling operators (e.g., Max-pooling) to effectively reduce the number of nodes while preserving the semantic information of the graph. Recently, to enhance GNNs in graph learning tasks, Manifold-Mixup, a data augmentation technique that produces synthetic graph data by linearly mixing a pair of graph data and their labels, has been widely adopted. However, the performance of Manifold-Mixup can be highly affected by graph pooling operators, and there have not been many studies that are dedicated to uncovering such affection. To bridge this gap, we take an early step to explore how graph pooling operators affect the performance of Mixup-based graph learning. To that end, we conduct a comprehensive empirical study by applying Manifold-Mixup to a formal characterization of graph pooling based on 11 graph pooling operations (9 hybrid pooling operators, 2 non-hybrid pooling operators). The experimental results on both natural language datasets (Gossipcop, Politifact) and programming language datasets (JAVA250, Python800) demonstrate that hybrid pooling operators are more effective for Manifold-Mixup than the standard Max-pooling and the state-of-the-art graph multiset transformer (GMT) pooling, in terms of producing more accurate and robust GNN models.

Salah Ghamizi, Maxime Cordy, Michail Papadakis et al.

Vulnerability to adversarial attacks is a well-known weakness of Deep Neural Networks. While most of the studies focus on natural images with standardized benchmarks like ImageNet and CIFAR, little research has considered real world applications, in particular in the medical domain. Our research shows that, contrary to previous claims, robustness of chest x-ray classification is much harder to evaluate and leads to very different assessments based on the dataset, the architecture and robustness metric. We argue that previous studies did not take into account the peculiarity of medical diagnosis, like the co-occurrence of diseases, the disagreement of labellers (domain experts), the threat model of the attacks and the risk implications for each successful attack. In this paper, we discuss the methodological foundations, review the pitfalls and best practices, and suggest new methodological considerations for evaluating the robustness of chest xray classification models. Our evaluation on 3 datasets, 7 models, and 18 diseases is the largest evaluation of robustness of chest x-ray classification models.

Thibault Simonetto, Salah Ghamizi, Antoine Desjardins et al.

State-of-the-art deep learning models for tabular data have recently achieved acceptable performance to be deployed in industrial settings. However, the robustness of these models remains scarcely explored. Contrary to computer vision, there is to date no realistic protocol to properly evaluate the adversarial robustness of deep tabular models due to intrinsic properties of tabular data such as categorical features, immutability, and feature relationship constraints. To fill this gap, we propose CAA, the first efficient evasion attack for constrained tabular deep learning models. CAA is an iterative parameter-free attack that combines gradient and search attacks to generate adversarial examples under constraints. We leverage CAA to build a benchmark of deep tabular models across three popular use cases: credit scoring, phishing and botnet attacks detection. Our benchmark supports ten threat models with increasing capabilities of the attacker, and reflects real-world attack scenarios for each use case. Overall, our results demonstrate how domain knowledge, adversarial training, and attack budgets impact the robustness assessment of deep tabular models and provide security practitioners with a set of recommendations to improve the robustness of deep tabular models against various evasion attack scenarios.

Martin Gubri, Maxime Cordy, Yves Le Traon

Transferability is the property of adversarial examples to be misclassified by other models than the surrogate model for which they were crafted. Previous research has shown that early stopping the training of the surrogate model substantially increases transferability. A common hypothesis to explain this is that deep neural networks (DNNs) first learn robust features, which are more generic, thus a better surrogate. Then, at later epochs, DNNs learn non-robust features, which are more brittle, hence worst surrogate. First, we tend to refute this hypothesis, using transferability as a proxy for representation similarity. We then establish links between transferability and the exploration of the loss landscape in parameter space, focusing on sharpness, which is affected by early stopping. This leads us to evaluate surrogate models trained with seven minimizers that minimize both loss value and loss sharpness. Among them, SAM consistently outperforms early stopping by up to 28.8 percentage points. We discover that the strong SAM regularization from large flat neighborhoods tightly links to transferability. Finally, the best sharpness-aware minimizers prove competitive with other training methods and complement existing transferability techniques.

Chaimaa Medjadji, Sadi Alawadi, Feras M. Awaysheh et al.

Federated Learning (FL) enables collaborative model training across decentralized clients while preserving data privacy by keeping raw data local. However, FL suffers from significant communication overhead due to the frequent exchange of high-dimensional model updates over constrained networks. In this paper, we present FedSparQ, a lightweight compression framework that dynamically sparsifies the gradient of each client through an adaptive threshold, applies half-precision quantization to retained entries and integrates residuals from error feedback to prevent loss of information. FedSparQ requires no manual tuning of sparsity rates or quantization schedules, adapts seamlessly to both homogeneous and heterogeneous data distributions, and is agnostic to model architecture. Through extensive empirical evaluation on vision benchmarks under independent and identically distributed (IID) and non-IID data, we show that FedSparQ substantially reduces communication overhead (reducing by 90% of bytes sent compared to FedAvg) while preserving or improving model accuracy (improving by 6% compared to FedAvg non-compressed solution or to state-of-the-art compression models) and enhancing convergence robustness (by 50%, compared to the other baselines). Our approach provides a practical, easy-to-deploy solution for bandwidth-constrained federated deployments and lays the groundwork for future extensions in adaptive precision and privacy-preserving protocols.

Yuejun Guo, Seifeddine Bettaieb, Qiang Hu et al.

Representing source code in a generic input format is crucial to automate software engineering tasks, e.g., applying machine learning algorithms to extract information. Visualizing code representations can further enable human experts to gain an intuitive insight into the code. Unfortunately, as of today, there is no universal tool that can simultaneously visualise different types of code representations. In this paper, we introduce a tool, CodeLens, which provides a visual interaction environment that supports various representation methods and helps developers understand and explore them. CodeLens is designed to support multiple programming languages, such as Java, Python, and JavaScript, and four types of code representations, including sequence of tokens, abstract syntax tree (AST), data flow graph (DFG), and control flow graph (CFG). By using CodeLens, developers can quickly visualize the specific code representation and also obtain the represented inputs for models of code. The Web-based interface of CodeLens is available at http://www.codelens.org. The demonstration video can be found at http://www.codelens.org/demo.

Chaimaa Medjadji, Sylvain Kubler, Yves Le Traon et al.

Federated Learning (FL) enables collaborative model training without centralizing data. However, real-world deployments must simultaneously address statistical heterogeneity across client data (non-IID), system heterogeneity in device capabilities, and communication efficiency. Existing FL approaches mitigate these challenges through improved aggregation, personalization, or knowledge distillation, but they almost universally assume a fixed client architecture, limiting adaptability to heterogeneous data complexity and hardware constraints. This architectural constraint often leads to suboptimal trade-offs between accuracy and efficiency in real-world FL systems. This work introduces FedKDNAS, a distillation-driven FL framework that combines client-side neural architecture selection with distillation of server-coordinated knowledge. Each client autonomously selects a lightweight model under accuracy-resource constraints. It then trains it locally using a hybrid objective combining supervised learning and knowledge distillation and shares only predictions on a public reference set. The server then aggregates and smooths these predictions, optionally combining them with a teacher model, to produce stable distillation targets for the next round. Extensive evaluation on six datasets against six representative FL baselines (FedAvg, Ditto, FedMD, FedDF, FedDistill, Local-KD) demonstrates that FedKDNAS consistently achieves superior Pareto efficiency, improving accuracy by up to 15\% under non-IID conditions, reducing client CPU usage by approximately 28\%, and decreasing communication overhead by up to 44 times while maintaining lightweight logit-based communication.

Xiongfei Wu, Mingfei Cheng, Xiaoning Ren et al.

Recent advances in foundation models (FMs), including large language models (LLMs), vision-language models (VLMs), and world models, have opened new opportunities for autonomous driving systems (ADSs) in perception, reasoning, decision-making, and interaction. However, ADSs are safety-critical cyber-physical systems, and integrating FMs into them raises substantial software engineering challenges in data curation, system design, deployment, evaluation, and assurance. To clarify this rapidly evolving landscape, we present an initial roadmap, grounded in a structured literature review, for integrating FMs into autonomous driving across three dimensions: FM infrastructure, in-vehicle integration, and practical deployment. For each dimension, we summarize the state of the art, identify key challenges, and highlight open research opportunities. Based on this analysis, we outline research directions for building reliable, safe, and trustworthy FM-enabled ADSs.

Chaimaa Medjadji, Guilain Leduc, Sylvain Kubler et al.

Federated Learning (FL) has emerged as a promising paradigm for collaborative model training across distributed edge devices while preserving data privacy especially with the huge increase amount of data due to the adoption of technologies which contributes to the growing number of IoT devices. Storing this amount of data centrally is challenging due to issues like limited communication, privacy, and regulations. FL can be Centralized (CFL), Decentralized (DFL), and Semi-decentralized (SDFL). Choosing the right FL architecture depends on the application's needs. However, very few research studies have experimentally compared these three types of architectures to not only understand the respective strengths and limitations, but also trade-offs between different performance indicators. This paper overcome this lack of analysis, conducting experimental analyses using the Fedstellar simulator, MNIST dataset, and MLP classifier.

Imane Hocine, Chaimaa Medjadji, Sylvain Kubler et al.

Federated Learning (FL) enables collaborative training across decentralized clients, but most methods assume aligned feature schemas, an assumption that rarely holds in tabular settings where clients observe only partially overlapping feature subsets. In these heterogeneous feature spaces, parameter-averaging methods (e.g., FedAvg) transfer little information across weakly overlapping or disjoint feature groups, limiting their effectiveness for federated imputation. To overcome this, we propose \textbf{FedHF-Impute}, a federated imputation framework that separates structural feature unavailability from conventional missingness and uses a shared global feature graph to propagate information across statistically related features through message passing. This enables indirect cross-client knowledge transfer, even when features are never jointly observed locally, while preserving standard federated communication. Under simulated partial schema overlap on the SECOM and AirQuality datasets, FedHF-Impute improves imputation accuracy (RMSE) over FL baselines by 26.9\%, and 8.4\% respectively, while achieving comparable performance on PhysioNET, with only a 0.3\% difference relative to the best baseline.

Yu Pan, Yang Hou, Xiongfei Wu et al.

Compositional speech-to-speech translation (S2ST) systems built upon speech large language models (SpeechLLMs) have recently shown promising performance. However, existing S2ST systems often either neglect source-language information or encode it through a language-as-label paradigm, representing each source language as an independent flat embedding. Such a design overlooks systematic linguistic structure shared across languages, which may limit data-efficient multilingual adaptation when supervised S2ST data are scarce. To address this issue, we propose S2ST-Omni 2, a many-to-one compositional S2ST framework that systematically reformulates multilingual language conditioning from flat language labels to structured typological priors. Specifically, S2ST-Omni 2 revisits language conditioning at three levels: typology-informed hierarchical language encoding for structured source-language representation, dynamically-gated language-aware Dual-CTC for content-adaptive acoustic modulation, and typology-aware LLM prompting for decoder-side linguistic guidance. Experiments on CVSS-C show that S2ST-Omni 2 achieves superior average performance among representative S2ST approaches across BLEU, COMET, ASR-BLEU, and BLASER 2.0 under the adopted evaluation protocol. Ablation studies indicate that the proposed representation-level, acoustic-level, and decoding-level strategies provide complementary benefits. Moreover, controlled data-budget analyses and a Japanese-to-English evaluation using only approximately 3 hours of supervised training data suggest that explicit typological priors provide useful inductive biases for data-efficient multilingual S2ST.

Milos Ojdanic, Ezekiel Soremekun, Renzo Degiovanni et al.

When software evolves, opportunities for introducing faults appear. Therefore, it is important to test the evolved program behaviors during each evolution cycle. We conduct an exploratory study to investigate the properties of commit-relevant mutants, i.e., the test elements of commit-aware mutation testing, by offering a general definition and an experimental approach to identify them. We thus, aim at investigating the prevalence, location, comparative advantages of commit-aware mutation testing over time (i.e., the program evolution) and the predictive power of several commit-related features to understand the essential properties for its best-effort application case. Our approach utilizes the impact of mutants and the effects of one mutant on another in capturing and analyzing the implicit interactions between the changed and unchanged code parts. The study analyses millions of mutants (over 10 million), 288 commits, five (5) different open-source software projects involving over 68,213 CPU days of computation and sets a ground truth where we perform our analysis. Our analysis shows that commit-relevant mutants are located mainly outside of program commit change (81%), while an effective selection of commit-relevant mutants can reduce the number of mutants by up to 93%. In addition, we demonstrate that commit relevant mutation testing is significantly more effective and efficient than state-of-the-art baselines. Our analysis of the predictive power of mutants and commit-related features in predicting commit-relevant mutants found that most proxy features do not reliably predict commit-relevant mutants. This empirical study highlights the properties of commit-relevant mutants and demonstrates the importance of identifying and selecting commit-relevant mutants when testing evolving software systems.

Renaud Rwemalika, Sarra Habchi, Mike Papadakis et al.

Test smells are known as bad development practices that reflect poor design and implementation choices in software tests. Over the last decade, test smells were heavily studied to measure their prevalence and impacts on test maintainability. However, these studies focused mainly on the unit level and to this day, the work on system tests that interact with the System Under Test through a Graphical User Interface remains limited. To fill the gap, we conduct an exploratory analysis of test smells occurring in System User Interactive Tests (SUIT). First, based on a multi-vocal literature review, we propose a catalog of 35 SUIT-specific smells. Then, we conduct an empirical analysis to assess the prevalence and refactoring of these smells in 48 industrial test suites and 12 open-source projects. We show that the same type of smells tends to appear in industrial and open-source projects, but the symptoms are not addressed in the same way. Smells such as Obscure Test, Sneaky Checking, and Over Checking show symptoms in more than 70% of the tests. Yet refactoring actions are much less frequent with less than 50% of the affected tests ever undergoing refactoring. Interestingly, while refactoring actions are rare, some smells, such as Narcissistic, disappear through the removal of old symptomatic tests and the introduction of new tests not presenting such symptoms.

Aayush Garg, Renzo Degiovanni, Matthieu Jimenez et al.

Vulnerability prediction refers to the problem of identifying system components that are most likely to be vulnerable. Typically, this problem is tackled by training binary classifiers on historical data. Unfortunately, recent research has shown that such approaches underperform due to the following two reasons: a) the imbalanced nature of the problem, and b) the inherently noisy historical data, i.e., most vulnerabilities are discovered much later than they are introduced. This misleads classifiers as they learn to recognize actual vulnerable components as non-vulnerable. To tackle these issues, we propose TROVON, a technique that learns from known vulnerable components rather than from vulnerable and non-vulnerable components, as typically performed. We perform this by contrasting the known vulnerable, and their respective fixed components. This way, TROVON manages to learn from the things we know, i.e., vulnerabilities, hence reducing the effects of noisy and unbalanced data. We evaluate TROVON by comparing it with existing techniques on three security-critical open source systems, i.e., Linux Kernel, OpenSSL, and Wireshark, with historical vulnerabilities that have been reported in the National Vulnerability Database (NVD). Our evaluation demonstrates that the prediction capability of TROVON significantly outperforms existing vulnerability prediction techniques such as Software Metrics, Imports, Function Calls, Text Mining, Devign, LSTM, and LSTM-RF with an improvement of 40.84% in Matthews Correlation Coefficient (MCC) score under Clean Training Data Settings, and an improvement of 35.52% under Realistic Training Data Settings.

Kui Liu, Shangwen Wang, Anil Koyuncu et al.

Test-based automated program repair has been a prolific field of research in software engineering in the last decade. Many approaches have indeed been proposed, which leverage test suites as a weak, but affordable, approximation to program specifications. Although the literature regularly sets new records on the number of benchmark bugs that can be fixed, several studies increasingly raise concerns about the limitations and biases of state-of-the-art approaches. For example, the correctness of generated patches has been questioned in a number of studies, while other researchers pointed out that evaluation schemes may be misleading with respect to the processing of fault localization results. Nevertheless, there is little work addressing the efficiency of patch generation, with regard to the practicality of program repair. In this paper, we fill this gap in the literature, by providing an extensive review on the efficiency of test suite based program repair. Our objective is to assess the number of generated patch candidates, since this information is correlated to (1) the strategy to traverse the search space efficiently in order to select sensical repair attempts, (2) the strategy to minimize the test effort for identifying a plausible patch, (3) as well as the strategy to prioritize the generation of a correct patch. To that end, we perform a large-scale empirical study on the efficiency, in terms of quantity of generated patch candidates of the 16 open-source repair tools for Java programs. The experiments are carefully conducted under the same fault localization configurations to limit biases.

Anil Koyuncu, Kui Liu, Tegawendé F. Bissyandé et al.

Patching is a common activity in software development. It is generally performed on a source code base to address bugs or add new functionalities. In this context, given the recurrence of bugs across projects, the associated similar patches can be leveraged to extract generic fix actions. While the literature includes various approaches leveraging similarity among patches to guide program repair, these approaches often do not yield fix patterns that are tractable and reusable as actionable input to APR systems. In this paper, we propose a systematic and automated approach to mining relevant and actionable fix patterns based on an iterative clustering strategy applied to atomic changes within patches. The goal of FixMiner is thus to infer separate and reusable fix patterns that can be leveraged in other patch generation systems. Our technique, FixMiner, leverages Rich Edit Script which is a specialized tree structure of the edit scripts that captures the AST-level context of the code changes. FixMiner uses different tree representations of Rich Edit Scripts for each round of clustering to identify similar changes. These are abstract syntax trees, edit actions trees, and code context trees. We have evaluated FixMiner on thousands of software patches collected from open source projects. Preliminary results show that we are able to mine accurate patterns, efficiently exploiting change information in Rich Edit Scripts. We further integrated the mined patterns to an automated program repair prototype, PARFixMiner, with which we are able to correctly fix 26 bugs of the Defects4J benchmark. Beyond this quantitative performance, we show that the mined fix patterns are sufficiently relevant to produce patches with a high probability of correctness: 81% of PARFixMiner's generated plausible patches are correct.

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.

Thomas Hartmann, Francois Fouquet, Yves Le Traon et al.

Intelligent software systems continuously analyze their surrounding environment and accordingly adapt their internal state. Depending on the criticality index of the situation, the system should dynamically focus or widen its analysis and reasoning scope. A naive -why have less when you can have more- approach would consist in systematically sampling the context at a very high rate and triggering the reasoning process regularly. This reasoning process would then need to mine a huge amount of data, extract a relevant view, and finally analyze this adequate view. This overall process would require some heavy resources and/or be time-consuming, conflicting with the (near) real-time response time requirements of intelligent systems. We claim that a continuous and more flexible navigation into context models, in space and in time, can significantly improve reasoning processes. This paper thus introduces a novel modeling approach together with a navigation concept, which consider time and locality as first-class properties crosscutting any model element, and enable the seamless navigation of models in this space-time continuum. In particular, we leverage a time-relative navigation (inspired by the space-time and distortion theory [7]) able to efficiently empower continuous reasoning processes. We integrate our approach into an open-source modeling framework and evaluate it on a smart grid reasoning engine for electric load prediction. We demonstrate that reasoners leveraging this distorted space-time continuum outperform the full sampling approach, and is compatible with most of (near) real-time requirements.

Amal AKLI, Mike PAPADAKIS, Maxime CORDY et al.

Large language models are increasingly used for code generation, yet the correctness of their outputs depends not only on model capability but also on how tasks are specified. Prior studies demonstrate that small changes in natural language prompts, particularly under-specification can substantially reduce code correctness; however, these findings are largely based on minimal-specification benchmarks such as HumanEval and MBPP, where limited structural redundancy may exaggerate sensitivity. In this exploratory study, we investigate how prompt structure, task complexity, and specification richness interact with LLM robustness to prompt mutations. We evaluate 10 different models across HumanEval and the structurally richer LiveCodeBench. Our results reveal that robustness is not a fixed property of LLMs but is highly dependent on prompt structure: the same under-specification mutations that degrade performance on HumanEval have near-zero net effect on LiveCodeBench due to redundancy across descriptions, constraints, examples, and I/O conventions. Surprisingly, we also find that prompt mutations can improve correctness. In LiveCodeBench, under-specification often breaks misleading lexical or structural cues that trigger incorrect retrieval-based solution strategies, leading to correctness improvements that counterbalance degradations. Manual analysis identifies consistent mechanisms behind these improvements, including the disruption of over-fitted terminology, removal of misleading constraints, and elimination of spurious identifier triggers. Overall, our study shows that structurally rich task descriptions can substantially mitigate the negative effects of under-specification and, in some cases, even enhance correctness. We outline categories of prompt modifications that positively influence the behavior of LLM code-generation, offering practical insights for writing robust prompts.

Amal Akli, Mike Papadakis, Maxime Cordy et al.

Large language models are widely used for code generation, yet they rely on an implicit assumption that the task descriptions are sufficiently detailed and well-formed. However, in practice, users may provide defective descriptions, which can have a strong effect on code correctness. To address this issue, we develop SpecValidator, a lightweight classifier based on a small model that has been parameter-efficiently finetuned, to automatically detect task description defects. We evaluate SpecValidator on three types of defects, Lexical Vagueness, Under-Specification and Syntax-Formatting on 3 benchmarks with task descriptions of varying structure and complexity. Our results show that SpecValidator achieves defect detection of F1 = 0.804 and MCC = 0.745, significantly outperforming GPT-5-mini (F1 = 0.469 and MCC = 0.281) and Claude Sonnet 4 (F1 = 0.518 and MCC = 0.359). Perhaps more importantly, our analysis indicates that SpecValidator can generalize to unseen issues and detect unknown Under-Specification defects in the original (real) descriptions of the benchmarks used. Our results also show that the robustness of LLMs in task description defects depends primarily on the type of defect and the characteristics of the task description, rather than the capacity of the model, with Under-Specification defects being the most severe. We further found that benchmarks with richer contextual grounding, such as LiveCodeBench, exhibit substantially greater resilience, highlighting the importance of structured task descriptions for reliable LLM-based code generation.

Maya Larbi, Amal Akli, Mike Papadakis et al.

Large Language Models (LLMs) have demonstrated impressive performance in code generation tasks under idealized conditions, where task descriptions are clear and precise. However, in practice, task descriptions frequently exhibit ambiguity, incompleteness, or internal contradictions. In this paper, we present the first empirical study examining the robustness of state-of-the-art code generation models when faced with such unclear task descriptions. We extend the HumanEval and MBPP benchmarks by systematically introducing realistic task descriptions flaws through guided mutation strategies, producing a dataset that mirrors the messiness of informal developer instructions. We evaluate multiple LLMs of varying sizes and architectures, analyzing their functional correctness and failure modes across task descriptions categories. Our findings reveal that even minor imperfections in task description phrasing can cause significant performance degradation, with contradictory task descriptions resulting in numerous logical errors. Moreover, while larger models tend to be more resilient than smaller variants, they are not immune to the challenges posed by unclear requirements. We further analyze semantic error patterns and identify correlations between description clarity, model behavior, and error types. Our results underscore the critical need for developing LLMs that are not only powerful but also robust to the imperfections inherent in natural user tasks, highlighting important considerations for improving model training strategies, designing more realistic evaluation benchmarks, and ensuring reliable deployment in practical software development environments.

Badr Souani, Ezekiel Soremekun, Mike Papadakis et al.

Large Language Models (LLMs) may portray discrimination towards certain individuals, especially those characterized by multiple attributes (aka intersectional bias). Discovering intersectional bias in LLMs is challenging, as it involves complex inputs on multiple attributes (e.g. race and gender). To address this challenge, we propose HInter, a test technique that synergistically combines mutation analysis, dependency parsing and metamorphic oracles to automatically detect intersectional bias in LLMs. HInter generates test inputs by systematically mutating sentences using multiple mutations, validates inputs via a dependency invariant and detects biases by checking the LLM response on the original and mutated sentences. We evaluate HInter using six LLM architectures and 18 LLM models (GPT3.5, Llama2, BERT, etc) and find that 14.61% of the inputs generated by HInter expose intersectional bias. Results also show that our dependency invariant reduces false positives (incorrect test inputs) by an order of magnitude. Finally, we observed that 16.62% of intersectional bias errors are hidden, meaning that their corresponding atomic cases do not trigger biases. Overall, this work emphasize the importance of testing LLMs for intersectional bias.

Fabien Bernier, Maxime Cordy, Yves Le Traon

Accurate electrical consumption forecasting is crucial for efficient energy management and resource allocation. While traditional time series forecasting relies on historical patterns and temporal dependencies, incorporating external factors -- such as weather indicators -- has shown significant potential for improving prediction accuracy in complex real-world applications. However, the inclusion of these additional features often degrades the performance of global predictive models trained on entire populations, despite improving individual household-level models. To address this challenge, we found that a hypernetwork architecture can effectively leverage external factors to enhance the accuracy of global electrical consumption forecasting models, by specifically adjusting the model weights to each consumer. We collected a comprehensive dataset spanning two years, comprising consumption data from over 6000 luxembourgish households and corresponding external factors such as weather indicators, holidays, and major local events. By comparing various forecasting models, we demonstrate that a hypernetwork approach outperforms existing methods when associated to external factors, reducing forecasting errors and achieving the best accuracy while maintaining the benefits of a global model.

Pantelis Dogoulis, Matthieu Jimenez, Salah Ghamizi et al.

This paper analyzes the robustness of state-of-the-art AI-based models for power grid operations under the $N-1$ security criterion. While these models perform well in regular grid settings, our results highlight a significant loss in accuracy following the disconnection of a line.%under this security criterion. Using graph theory-based analysis, we demonstrate the impact of node connectivity on this loss. Our findings emphasize the need for practical scenario considerations in developing AI methodologies for critical infrastructure.

Jie Zhang, Wei Ma, Qiang Hu et al.

Existing methods for generating adversarial code examples face several challenges: limted availability of substitute variables, high verification costs for these substitutes, and the creation of adversarial samples with noticeable perturbations. To address these concerns, our proposed approach, RNNS, uses a search seed based on historical attacks to find potential adversarial substitutes. Rather than directly using the discrete substitutes, they are mapped to a continuous vector space using a pre-trained variable name encoder. Based on the vector representation, RNNS predicts and selects better substitutes for attacks. We evaluated the performance of RNNS across six coding tasks encompassing three programming languages: Java, Python, and C. We employed three pre-trained code models (CodeBERT, GraphCodeBERT, and CodeT5) that resulted in a cumulative of 18 victim models. The results demonstrate that RNNS outperforms baselines in terms of ASR and QT. Furthermore, the perturbation of adversarial examples introduced by RNNS is smaller compared to the baselines in terms of the number of replaced variables and the change in variable length. Lastly, our experiments indicate that RNNS is efficient in attacking defended models and can be employed for adversarial training.

Sai Sathiesh Rajan, Ezekiel Soremekun, Yves Le Traon et al.

Ensuring that all classes of objects are detected with equal accuracy is essential in AI systems. For instance, being unable to identify any one class of objects could have fatal consequences in autonomous driving systems. Hence, ensuring the reliability of image recognition systems is crucial. This work addresses how to validate group fairness in image recognition software. We propose a distribution-aware fairness testing approach (called DistroFair) that systematically exposes class-level fairness violations in image classifiers via a synergistic combination of out-of-distribution (OOD) testing and semantic-preserving image mutation. DistroFair automatically learns the distribution (e.g., number/orientation) of objects in a set of images. Then it systematically mutates objects in the images to become OOD using three semantic-preserving image mutations - object deletion, object insertion and object rotation. We evaluate DistroFair using two well-known datasets (CityScapes and MS-COCO) and three major, commercial image recognition software (namely, Amazon Rekognition, Google Cloud Vision and Azure Computer Vision). Results show that about 21% of images generated by DistroFair reveal class-level fairness violations using either ground truth or metamorphic oracles. DistroFair is up to 2.3x more effective than two main baselines, i.e., (a) an approach which focuses on generating images only within the distribution (ID) and (b) fairness analysis using only the original image dataset. We further observed that DistroFair is efficient, it generates 460 images per hour, on average. Finally, we evaluate the semantic validity of our approach via a user study with 81 participants, using 30 real images and 30 corresponding mutated images generated by DistroFair. We found that images generated by DistroFair are 80% as realistic as real-world images.

Salijona Dyrmishi, Salah Ghamizi, Thibault Simonetto et al.

While the literature on security attacks and defense of Machine Learning (ML) systems mostly focuses on unrealistic adversarial examples, recent research has raised concern about the under-explored field of realistic adversarial attacks and their implications on the robustness of real-world systems. Our paper paves the way for a better understanding of adversarial robustness against realistic attacks and makes two major contributions. First, we conduct a study on three real-world use cases (text classification, botnet detection, malware detection)) and five datasets in order to evaluate whether unrealistic adversarial examples can be used to protect models against realistic examples. Our results reveal discrepancies across the use cases, where unrealistic examples can either be as effective as the realistic ones or may offer only limited improvement. Second, to explain these results, we analyze the latent representation of the adversarial examples generated with realistic and unrealistic attacks. We shed light on the patterns that discriminate which unrealistic examples can be used for effective hardening. We release our code, datasets and models to support future research in exploring how to reduce the gap between unrealistic and realistic adversarial attacks.

Milos Ojdanic, Aayush Garg, Ahmed Khanfir et al.

Fault seeding is typically used in controlled studies to evaluate and compare test techniques. Central to these techniques lies the hypothesis that artificially seeded faults involve some form of realistic properties and thus provide realistic experimental results. In an attempt to strengthen realism, a recent line of research uses advanced machine learning techniques, such as deep learning and Natural Language Processing (NLP), to seed faults that look like (syntactically) real ones, implying that fault realism is related to syntactic similarity. This raises the question of whether seeding syntactically similar faults indeed results in semantically similar faults and more generally whether syntactically dissimilar faults are far away (semantically) from the real ones. We answer this question by employing 4 fault-seeding techniques (PiTest - a popular mutation testing tool, IBIR - a tool with manually crafted fault patterns, DeepMutation - a learning-based fault seeded framework and CodeBERT - a novel mutation testing tool that use code embeddings) and demonstrate that syntactic similarity does not reflect semantic similarity. We also show that 60%, 47%, 43%, and 7% of the real faults of Defects4J V2 are semantically resembled by CodeBERT, PiTest, IBIR, and DeepMutation faults. We then perform an objective comparison between the techniques and find that CodeBERT and PiTest have similar fault detection capabilities that subsume IBIR and DeepMutation, and that IBIR is the most cost-effective technique. Moreover, the overall fault detection of PiTest, CodeBERT, IBIR, and DeepMutation was, on average, 54%, 53%, 37%, and 7%.

Aayush Garg, Milos Ojdanic, Renzo Degiovanni et al.

Mutation testing research has indicated that a major part of its application cost is due to the large number of low utility mutants that it introduces. Although previous research has identified this issue, no previous study has proposed any effective solution to the problem. Thus, it remains unclear how to mutate and test a given piece of code in a best effort way, i.e., achieving a good trade-off between invested effort and test effectiveness. To achieve this, we propose Cerebro, a machine learning approach that statically selects subsuming mutants, i.e., the set of mutants that resides on the top of the subsumption hierarchy, based on the mutants' surrounding code context. We evaluate Cerebro using 48 and 10 programs written in C and Java, respectively, and demonstrate that it preserves the mutation testing benefits while limiting application cost, i.e., reduces all cost application factors such as equivalent mutants, mutant executions, and the mutants requiring analysis. We demonstrate that Cerebro has strong inter-project prediction ability, which is significantly higher than two baseline methods, i.e., supervised learning on features proposed by state-of-the-art, and random mutant selection. More importantly, our results show that Cerebro's selected mutants lead to strong tests that are respectively capable of killing 2 times higher than the number of subsuming mutants killed by the baselines when selecting the same number of mutants. At the same time, Cerebro reduces the cost-related factors, as it selects, on average, 68% fewer equivalent mutants, while requiring 90% fewer test executions than the baselines.

Sarra Habchi, Guillaume Haben, Mike Papadakis et al.

Test flakiness forms a major testing concern. Flaky tests manifest non-deterministic outcomes that cripple continuous integration and lead developers to investigate false alerts. Industrial reports indicate that on a large scale, the accrual of flaky tests breaks the trust in test suites and entails significant computational cost. To alleviate this, practitioners are constrained to identify flaky tests and investigate their impact. To shed light on such mitigation mechanisms, we interview 14 practitioners with the aim to identify (i) the sources of flakiness within the testing ecosystem, (ii) the impacts of flakiness, (iii) the measures adopted by practitioners when addressing flakiness, and (iv) the automation opportunities for these measures. Our analysis shows that, besides the tests and code, flakiness stems from interactions between the system components, the testing infrastructure, and external factors. We also highlight the impact of flakiness on testing practices and product quality and show that the adoption of guidelines together with a stable infrastructure are key measures in mitigating the problem.

Yuejun Guo, Qiang Hu, Maxime Cordy et al.

Active learning is an established technique to reduce the labeling cost to build high-quality machine learning models. A core component of active learning is the acquisition function that determines which data should be selected to annotate. State-of-the-art acquisition functions -- and more largely, active learning techniques -- have been designed to maximize the clean performance (e.g. accuracy) and have disregarded robustness, an important quality property that has received increasing attention. Active learning, therefore, produces models that are accurate but not robust. In this paper, we propose \emph{robust active learning}, an active learning process that integrates adversarial training -- the most established method to produce robust models. Via an empirical study on 11 acquisition functions, 4 datasets, 6 DNN architectures, and 15105 trained DNNs, we show that robust active learning can produce models with the robustness (accuracy on adversarial examples) ranging from 2.35\% to 63.85\%, whereas standard active learning systematically achieves negligible robustness (less than 0.20\%). Our study also reveals, however, that the acquisition functions that perform well on accuracy are worse than random sampling when it comes to robustness. We, therefore, examine the reasons behind this and devise a new acquisition function that targets both clean performance and robustness. Our acquisition function -- named density-based robust sampling with entropy (DRE) -- outperforms the other acquisition functions (including random) in terms of robustness by up to 24.40\% (3.84\% than random particularly), while remaining competitive on accuracy. Additionally, we prove that DRE is applicable as a test selection metric for model retraining and stands out from all compared functions by up to 8.21\% robustness.

Wei Ma, Mengjie Zhao, Ezekiel Soremekun et al.

Code embedding is a keystone in the application of machine learning on several Software Engineering (SE) tasks. To effectively support a plethora of SE tasks, the embedding needs to capture program syntax and semantics in a way that is generic. To this end, we propose the first self-supervised pre-training approach (called GraphCode2Vec) which produces task-agnostic embedding of lexical and program dependence features. GraphCode2Vec achieves this via a synergistic combination of code analysis and Graph Neural Networks. GraphCode2Vec is generic, it allows pre-training, and it is applicable to several SE downstream tasks. We evaluate the effectiveness of GraphCode2Vec on four (4) tasks (method name prediction, solution classification, mutation testing and overfitted patch classification), and compare it with four (4) similarly generic code embedding baselines (Code2Seq, Code2Vec, CodeBERT, GraphCodeBERT) and 7 task-specific, learning-based methods. In particular, GraphCode2Vec is more effective than both generic and task-specific learning-based baselines. It is also complementary and comparable to GraphCodeBERT (a larger and more complex model). We also demonstrate through a probing and ablation study that GraphCode2Vec learns lexical and program dependence features and that self-supervised pre-training improves effectiveness.

Thibault Simonetto, Salijona Dyrmishi, Salah Ghamizi et al.

The generation of feasible adversarial examples is necessary for properly assessing models that work in constrained feature space. However, it remains a challenging task to enforce constraints into attacks that were designed for computer vision. We propose a unified framework to generate feasible adversarial examples that satisfy given domain constraints. Our framework can handle both linear and non-linear constraints. We instantiate our framework into two algorithms: a gradient-based attack that introduces constraints in the loss function to maximize, and a multi-objective search algorithm that aims for misclassification, perturbation minimization, and constraint satisfaction. We show that our approach is effective in four different domains, with a success rate of up to 100%, where state-of-the-art attacks fail to generate a single feasible example. In addition to adversarial retraining, we propose to introduce engineered non-convex constraints to improve model adversarial robustness. We demonstrate that this new defense is as effective as adversarial retraining. Our framework forms the starting point for research on constrained adversarial attacks and provides relevant baselines and datasets that future research can exploit.

Guillaume Haben, Sarra Habchi, Mike Papadakis et al.

Flakiness is a major concern in Software testing. Flaky tests pass and fail for the same version of a program and mislead developers who spend time and resources investigating test failures only to discover that they are false alerts. In practice, the defacto approach to address this concern is to rerun failing tests hoping that they would pass and manifest as false alerts. Nonetheless, completely filtering out false alerts may require a disproportionate number of reruns, and thus incurs important costs both computation and time-wise. As an alternative to reruns, we propose Fair, a novel, lightweight approach that classifies test failures into false alerts and legitimate failures. Fair relies on a classifier and a set of features from the failures and test artefacts. To build and evaluate our machine learning classifier, we use the continuous integration of the Chromium project. In particular, we collect the properties and artefacts of more than 1 million test failures from 2,000 builds. Our results show that Fair can accurately distinguish legitimate failures from false alerts, with an MCC up to 95%. Moreover, by studying different test categories: GUI, integration and unit tests, we show that Fair classifies failures accurately even when the number of failures is limited. Finally, we compare the costs of our approach to reruns and show that Fair could save up to 20 minutes of computation time per build.

Salah Ghamizi, Maxime Cordy, Mike Papadakis et al.

Vulnerability to adversarial attacks is a well-known weakness of Deep Neural networks. While most of the studies focus on single-task neural networks with computer vision datasets, very little research has considered complex multi-task models that are common in real applications. In this paper, we evaluate the design choices that impact the robustness of multi-task deep learning networks. We provide evidence that blindly adding auxiliary tasks, or weighing the tasks provides a false sense of robustness. Thereby, we tone down the claim made by previous research and study the different factors which may affect robustness. In particular, we show that the choice of the task to incorporate in the loss function are important factors that can be leveraged to yield more robust models.

Yuejun Guo, Qiang Hu, Maxime Cordy et al.

Deep Neural Networks (DNNs) are vulnerable to adversarial examples, which causes serious threats to security-critical applications. This motivated much research on providing mechanisms to make models more robust against adversarial attacks. Unfortunately, most of these defenses, such as gradient masking, are easily overcome through different attack means. In this paper, we propose MUTEN, a low-cost method to improve the success rate of well-known attacks against gradient-masking models. Our idea is to apply the attacks on an ensemble model which is built by mutating the original model elements after training. As we found out that mutant diversity is a key factor in improving success rate, we design a greedy algorithm for generating diverse mutants efficiently. Experimental results on MNIST, SVHN, and CIFAR10 show that MUTEN can increase the success rate of four attacks by up to 0.45.

Sarra Habchi, Maxime Cordy, Mike Papadakis et al.

Background: Test flakiness is identified as a major issue that compromises the regression testing process of complex software systems. Flaky tests manifest non-deterministic behaviour, send confusing signals to developers, and break their trust in test suites. Both industrial reports and research studies highlighted the negative impact of flakiness on software quality and developers' productivity. While researchers strive to devise solutions that could help developers addressing test flakiness, the elaboration and assessment of these solutions are hindered by the lack of datasets large enough to leverage learning techniques. Aim: To address this lack, we conduct an exploratory study that investigates a new mean for producing datasets of flaky tests. Method: We propose an approach that relies on program mutation to inject flakiness in software tests. In particular, we plan to delete helper statements from tests to make their outcomes order-dependent, i.e., pass in certain running orders but fail in other orders. We intend to apply our mutation-based approach to a set of 14 Java projects to assess the effectiveness of test mutation in injecting order-dependency and generate a new dataset that could be used to study flakiness.

Adriano Franci, Maxime Cordy, Martin Gubri et al.

Graph-based Semi-Supervised Learning (GSSL) is a practical solution to learn from a limited amount of labelled data together with a vast amount of unlabelled data. However, due to their reliance on the known labels to infer the unknown labels, these algorithms are sensitive to data quality. It is therefore essential to study the potential threats related to the labelled data, more specifically, label poisoning. In this paper, we propose a novel data poisoning method which efficiently approximates the result of label inference to identify the inputs which, if poisoned, would produce the highest number of incorrectly inferred labels. We extensively evaluate our approach on three classification problems under 24 different experimental settings each. Compared to the state of the art, our influence-driven attack produces an average increase of error rate 50\% higher, while being faster by multiple orders of magnitude. Moreover, our method can inform engineers of inputs that deserve investigation (relabelling them) before training the learning model. We show that relabelling one-third of the poisoned inputs (selected based on their influence) reduces the poisoning effect by 50\%.

Ahmed Khanfir, Anil Koyuncu, Mike Papadakis et al.

Much research on software engineering and software testing relies on experimental studies based on fault injection. Fault injection, however, is not often relevant to emulate real-world software faults since it "blindly" injects large numbers of faults. It remains indeed challenging to inject few but realistic faults that target a particular functionality in a program. In this work, we introduce IBIR, a fault injection tool that addresses this challenge by exploring change patterns associated to user-reported faults. To inject realistic faults, we create mutants by retargeting a bug report driven automated program repair system, i.e., reversing its code transformation templates. IBIR is further appealing in practice since it requires deep knowledge of neither of the code nor the tests, but just of the program's relevant bug reports. Thus, our approach focuses the fault injection on the feature targeted by the bug report. We assess IBIR by considering the Defects4J dataset. Experimental results show that our approach outperforms the fault injection performed by traditional mutation testing in terms of semantic similarity with the original bug, when applied at either system or class levels of granularity, and provides better, statistically significant, estimations of test effectiveness (fault detection). Additionally, when injecting 100 faults, IBIR injects faults that couple with the real ones in 36% of the cases, while mutants from mutation testing inject less than 1%. Overall, IBIR targets real functionality and injects realistic and diverse faults.