Search Results (49)

The rise of capture-the-flag (CTF) competitions and offensive security training requires the deployment of systems that are, by design, flawed. This creates a unique architectural paradox: how does one host a system intended to be compromised without compromising the host itself? This paper classifies the security principles of “range engineering”—the discipline of engineering the environment. This research study synthesizes evidence across the cyber-range, honeypot, ICS/OT testbed, and cloud-isolation literature to derive a containment-focused classification of threat planes, security invariants, boundary mechanisms and properties, and operational controls for intentionally vulnerable environments used in education, training, and research. Five security invariants are derived under the assumption of expected compromise and mapped to boundary families and measurable operational objectives. The analysis further identifies under-evidenced areas, particularly control-plane isolation, corrective controls for cross-tenant failures, and systematic validation of externalization defenses. Full article

This paper presents a Systematization of Knowledge (SoK) on the evaluation methodologies and capability boundaries of Large Language Models (LLMs) in cybersecurity. We propose a Three-Dimensional Taxonomy Matrix to systematize existing metrics across offensive domains, defensive applications, and inherent architectural flaws. Beyond categorization, this matrix functions as a predictive framework to expose structural evaluation blind spots. Specifically, by intersecting target domains with failure attributions, it identifies a critical, unresolved frontier: measuring cross-architecture semantic equivalence in low-level reverse engineering. Empirically, synthesizing 39 frontier benchmarks reveals a systemic evaluation gap: static metric success rarely translates into end-to-end adversarial efficacy. In offensive domains, high penetration rates correlate strongly with pre-training data contamination. When subjected to semantics-preserving code obfuscation as a stress test, zero-shot, tool-free exploit success rates collapse to near 0%. In defensive contexts, cross-procedural code auditing struggles, yielding a peak F1-score of only 23.83%. Furthermore, models suffer from over-alignment-induced functional degradation, with joint-testing frameworks recording up to a 77% functional loss in automated program repair. Our analysis strongly suggests that purely autoregressive mechanisms drive severe technical hallucinations, evidenced by a 19.7% package dependency fabrication rate. Evaluations also expose significant attack surfaces and a significant safety-utility tradeoff: models succumb to prompt leakage attacks at rates up to 86.2%, while heavily aligned versions simultaneously exhibit excessively high False Refusal Rates (FRR) for benign, borderline security queries. Finally, we delineate a theoretical neuro-symbolic roadmap—integrating LLM heuristics with deterministic formal methods—to structurally mitigate the limitations of the autoregressive paradigm. Full article

In modern cybersecurity and deep learning, conditional password guessing plays a critical role in improving password-cracking efficiency by leveraging known patterns and constraints. In contrast with traditional brute-force or dictionary-based attacks, we developed an approach that adopts a latent diffusion model to simulate human password selection behavior, generating more realistic password candidates. We incorporated masked character inputs as conditions and applied advanced feature extraction to capture common patterns such as character substitutions and typing habits. Furthermore, we employed visualization techniques, including autoencoders and principal component analysis, to analyze password distributions, enhancing model interpretability and aiding both offensive and defensive security strategies. Full article

Cybersecurity education requires practical activities such as malware analysis, phishing detection, and Capture the Flag (CTF) challenges. These exercises enable students to actively apply theoretical concepts in realistic scenarios, fostering experiential learning. This article introduces an innovative pedagogical approach relying on gamification in cybersecurity courses, combining technical problem-solving with human factors such as social engineering and risk-taking behavior. By integrating interactive challenges into the courses, engagement and motivation have been enhanced, while addressing both technological and managerial dimensions of cybersecurity. Observations from course implementation indicate that students demonstrate higher involvement when participating in supervised offensive security tasks and social engineering simulations within controlled environments. These findings highlight the potential of gamified strategies to strengthen cybersecurity competencies and promote ethical awareness, paving the way for future research on long-term cybersecurity learning outcomes. Full article

Urban bail systems rely heavily on cash bonds to secure the pretrial release of criminal defendants awaiting trial, despite longstanding criticism that this practice disproportionately incarcerates indigent defendants solely because they cannot afford to pay. Many large jurisdictions employ bail bond schedules that assign monetary amounts based primarily on the seriousness of the offense. Using data on 5322 defendants released on bail in 35 large urban counties, this study examines whether bail amount or a defendant’s prior criminal history better predicts pretrial failure, defined as pretrial rearrest and failure to appear. Results show that prior criminal history is a substantially stronger and more consistent predictor of pretrial failure than bail amount. Bail amount also exhibits no meaningful association with rearrest and only a modest relationship with failure to appear. These findings suggest that community safety may be better served by substantially reducing reliance on cash bail and placing greater emphasis on prior criminal history in pretrial release decisions. Full article

The convergence of ubiquitous connectivity, large-scale data generation, and rapid advancements in machine learning is transforming the field of cybersecurity. The widespread adoption of interconnected systems including Internet of Things devices, mobile platforms, and cloud infrastructures has introduced new attack surfaces and significantly increased the complexity of securing digital environments. Concurrently, these technologies have enabled the development of intelligent, data-driven defense strategies. Achieving effective protection in these settings requires not only applying machine learning to detect and prevent threats but also recognizing that such models can themselves become targets of adversarial manipulation. This survey presents a comprehensive analysis of recent progress at the intersection of machine learning and cybersecurity. It explores defensive applications such as malware detection, network traffic classification, and anomaly detection, as well as offensive strategies including adversarial evasion, poisoning, and backdoor attacks. Particular attention is paid to adversarial machine learning, highlighting the increasing sophistication of attacks that exploit model vulnerabilities and the corresponding evolution of defense mechanisms. Beyond synthesizing current research, the survey also identifies key open challenges and emerging research directions. This survey provides a comprehensive and accessible reference for researchers and practitioners aiming to understand and advance the secure application of machine learning across diverse cybersecurity domains. Full article

The study of attack–defense game decision making in critical infrastructure systems confronting intelligent adversaries, grounded in complex network theory, has emerged as a prominent topic in the field of network security. Most existing research centers on game-theoretic analysis under conditions of complete information and assumes that the attacker and defender share congruent criteria for evaluating target values. However, in reality, asymmetric value perception may lead to different evaluation criteria for both the offensive and defensive sides. This paper examines the game problem wherein the attacker and defender possess distinct target value evaluation criteria. The research findings reveal that both the attacker and defender have their own “advantage ranges” for value assessment, and topological heterogeneity is the reason for this phenomenon. Within their respective advantage ranges, the attacker or defender can adopt clear-cut strategies to secure optimal benefits—without needing to consider their opponents’ decisions. Outside these ranges, we explore how the attacker can leverage small-sample detection outcomes to probabilistically infer defenders’ strategies, and we further analyze the attackers’ preference strategy selections under varying acceptable security thresholds and penalty coefficients. The research results deliver more practical solutions for games involving uncertain value criteria. Full article

This observational retrospective study aimed to examine the association between team behaviour during periods of one-goal leads and subsequent match outcomes while accounting for team level and match location. All matches (n = 380) of the English Premier League (EPL) during the season 2023/24 were analyzed. Team behaviour was evaluated by shots every 10 min during a one-goal lead (SP10MDOGL), a time-normalized indicator of offensive activity that reflects a team’s strategic orientation while protecting a narrow lead. Mixed effects multinomial logistic regression was used to establish the association between SP10MDOGL and the match outcome. Results indicated that increased SP10MDOGL was strongly associated with a higher likelihood of both drawing (Odds ratio (OR) = 2.37, 95% confidence interval (CI) = 1.29–4.33; Cohen’s d (d) = 0.47) and winning (OR = 3.38; 95%CI = 1.93–5.92; d = 0.67) compared to losing. This association remained consistent across high-, intermediate-, and low-level teams regardless of whether they played at home or away. These findings suggest that maintaining an offensive approach through an increased number of shots during a one-goal lead is associated with a higher likelihood of securing positive match outcomes within the elite-level football context, such as the EPL. Soccer coaches should consider implementing proactive offensive strategies when protecting a narrow lead, regardless of their team level and match location. Full article

The industrialization of cybercrime, principally through Malware-as-a-Service (MaaS), has elevated HTTP cookie theft to a critical cybersecurity challenge, enabling attackers to bypass multi-factor authentication and perpetrate large-scale account takeovers. Employing a Holistic and Integrative Review methodology, this paper dissects the intricate, adaptive ecosystem of MaaS-driven cookie theft. We systematically characterize the co-evolving arms race between offensive and defensive strategies (2020–2025), revealing a critical strategic asymmetry where attackers optimize for speed and low cost, while effective defenses demand significant resources. To shift security from a reactive to an anticipatory posture, a multi-dimensional predictive framework is not only proposed but is also detailed as a formalized, testable algorithm, integrating technical, economic, and behavioral indicators to forecast emerging threat trajectories. Our findings conclude that long-term security hinges on disrupting the underlying cybercriminal economic model; we therefore reframe proactive countermeasures like Zero-Trust principles and ephemeral tokens as economic weapons designed to devalue the stolen asset. Finally, the paper provides a prioritized, multi-year research roadmap and a practical decision-tree framework to guide the implementation of these advanced, collaborative cybersecurity strategies to counter this pervasive and evolving threat. Full article

Henri Lefebvre’s “right to the city” has rarely been examined through an intersectional feminist lens, leaving unnoticed the uneven burdens that urban design and policy place on women. This article bridges that gap by combining constitutional analysis, survey data (n = 736), in-depth interviews, and participatory observation to assess how Quito’s public spaces affect women’s safety and mobility. Quantitative results show that 81% of respondents endured sexual or offensive remarks, 69.8% endured obscene gestures, and 38% endured severe harassment in the month before the survey; 43% of these incidents occurred only days or weeks beforehand, underscoring their routine nature. Qualitative narratives reveal behavioral adaptations—altered routes, self-policing dress codes, and distrust of authorities—and identify poorly lit corridors and weak institutional presence as spatial amplifiers of violence. Analysis of Quito’s “Safe City” program exposes a gulf between its ambitious rhetoric and its narrow, transport-centered implementation. We conclude that constitutional guarantees of participation, appropriation, and urban life will remain aspirational until urban planning mainstreams gender-sensitive design, secures intersectoral resources, and embeds women’s substantive participation throughout policy cycles. A feminist reimagining of Quito’s public realm is therefore indispensable to transform the right to the city from legal principle into lived reality. Full article

The detection of abusive language in Roman Urdu is important for secure digital interaction. This work investigates machine learning (ML), deep learning (DL), and transformer-based methods for detecting offensive language in Roman Urdu comments collected from YouTube news channels. Extracted features use TF-IDF and Count Vectorizer for unigrams, bigrams, and trigrams. Of all the ML models—Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), and Naïve Bayes (NB)—the best performance was achieved by the same SVM. DL models involved evaluating Bi-LSTM and CNN models, where the CNN model outperformed the others. Moreover, transformer variants such as LLaMA 2 and ModernBERT (MBERT) were instantiated and fine-tuned with LoRA (Low-Rank Adaptation) for better efficiency. LoRA has been tuned for large language models (LLMs), a family of advanced machine learning frameworks, based on the principle of making the process efficient with extremely low computational cost with better enhancement. According to the experimental results, LLaMA 2 with LoRA attained the highest F1-score of 96.58%, greatly exceeding the performance of other approaches. To elaborate, LoRA-optimized transformers perform well in capturing detailed subtleties of linguistic nuances, lending themselves well to Roman Urdu offensive language detection. The study compares the performance of conventional and contemporary NLP methods, highlighting the relevance of effective fine-tuning methods. Our findings pave the way for scalable and accurate automated moderation systems for online platforms supporting multiple languages. Full article

Traditional command and control (C2) frameworks struggle with evasion, automation, and resilience against modern detection techniques. This paper introduces covert C2 (C3), a novel C2 framework designed to enhance operational security and minimize detection. C3 employs a decentralized architecture, enabling independent victim communication with the C2 server for covert persistence. Its adaptable design supports diverse post-exploitation and lateral movement techniques for optimized results across various environments. Through optimized performance and the use of the native messaging API, C3 agents achieve a demonstrably low detection rate against prevalent Endpoint Detection and Response (EDR) solutions. A publicly available proof-of-concept implementation demonstrates C3’s effectiveness in real-world adversarial simulations, specifically in direct code execution for privilege escalation and lateral movement. Our findings indicate that integrating novel techniques, such as the native messaging API, and a decentralized architecture significantly improves the stealth, efficiency, and reliability of offensive operations. The paper further analyzes C3’s post-exploitation behavior, explores relevant defense strategies, and compares it with existing C2 solutions, offering practical insights for enhancing network security. Full article

Conducting penetration testing (pentesting) in cybersecurity is a crucial turning point for identifying vulnerabilities within the framework of Information Technology (IT), where real malicious offensive behavior is simulated to identify potential weaknesses and strengthen preventive controls. Given the complexity of the tests, time constraints, and the specialized level of expertise required for pentesting, analysis and exploitation tools are commonly used. Although useful, these tools often introduce uncertainty in findings, resulting in high rates of false positives. To enhance the effectiveness of these tests, Machine Learning (ML) has been integrated, showing significant potential for identifying anomalies across various security areas through detailed detection of underlying malicious patterns. However, pentesting environments are unpredictable and intricate, requiring analysts to make extensive efforts to understand, explore, and exploit them. This study considers these challenges, proposing a recommendation system based on a context-rich, vocabulary-aware transformer capable of processing questions related to the target environment and offering responses based on necessary pentest batteries evaluated by a Reinforcement Learning (RL) estimator. This RL component assesses optimal attack strategies based on previously learned data and dynamically explores additional attack vectors. The system achieved an F1 score and an Exact Match rate over $97.0$%, demonstrating its accuracy and effectiveness in selecting relevant pentesting strategies. Full article

Adversarial attacks expose the latent vulnerabilities within artificial intelligence systems, necessitating a reassessment and enhancement of model robustness to ensure the reliability and security of deep learning models against malicious attacks. We propose a fast method designed to efficiently find sample points close to the decision boundary. By computing the gradient information of each class in the input samples and comparing these gradient differences with the true class, we can identify the target class most sensitive to the decision boundary, thus generating adversarial examples. This technique is referred to as the “You Only Attack Once” (YOAO) algorithm. Compared to the DeepFool algorithm, this method requires only a single iteration to achieve effective attack results. The experimental results demonstrate that the proposed algorithm outperforms the original approach in various scenarios, especially in resource-constrained environments. Under a single iteration, it achieves a 70.6% higher success rate of the attacks compared to the DeepFool algorithm. Our proposed method shows promise for widespread application in both offensive and defensive strategies for diverse deep learning models. We investigated the relationship between classifier accuracy and adversarial attack success rate, comparing the algorithm with others. Our experiments validated that the proposed algorithm exhibits higher attack success rates and efficiency. Furthermore, we performed data visualization on the ImageNet dataset, demonstrating that the proposed algorithm focuses more on attacking important features. Finally, we discussed the existing issues with the algorithm and outlined future research directions. Our code will be made public upon acceptance of the paper. Full article

In order to analyze the choice of the optimal strategy of cyber security attack and defense in the unmanned aerial vehicles’ (UAVs) cyber range, a game model-based UAV cyber range risk assessment method is constructed. Through the attack and defense tree model, the risk assessment method is calculated. The model of attack and defense game with incomplete information is established and the Bayesian–Nash equilibrium of mixed strategy is calculated. The model and method focus on the mutual influence of the actions of both sides and the dynamic change in the confrontation process. According to the calculation methods of different benefits of different strategies selected in the offensive and defensive game, the risk assessment and calculation of the UAV cyber range are carried out based on the probability distribution of the defender’s benefits and the attacker’s optimal strategy selection. An example is given to prove the feasibility and effectiveness of this method. Full article