Papers

Device-free crowd-counting using WiFi Channel State Information (CSI) is a key enabling technology for a new generation of privacy-preserving Internet of Things (IoT) applications. However, practical deployment is severely hampered by the domain shift problem, where models trained in one environment fail to generalise to another. To overcome this, we propose a novel two-stage framework centred on a CSI-ResNet-A architecture. This model is pre-trained via self-supervised contrastive learning to learn domain-invariant representations and leverages lightweight Adapter modules for highly efficient fine-tuning. The resulting event sequence is then processed by a stateful counting machine to produce a final, stable occupancy estimate. We validate our framework extensively. On our WiFlow dataset, our unsupervised approach excels in a 10-shot learning scenario, achieving a final Mean Absolute Error (MAE) of just 0.44–a task where supervised baselines fail. To formally quantify robustness, we introduce the Generalisation Index (GI), on which our model scores near-perfectly, confirming its ability to generalise. Furthermore, our framework sets a new state-of-the-art public WiAR benchmark with 98.8% accuracy. Our ablation studies reveal the core strength of our design: adapter-based fine-tuning achieves performance within 1% of a full fine-tune (98.84% vs. 99.67%) while training 97.2% fewer parameters. Our work provides a practical and scalable solution for developing robust sensing systems ready for real-world IoT deployments.

WiFi Channel State Information (CSI) has shown promise for single-person gait identification, raising interest in its use for contactless biometrics, continuous authentication, and passive identification. However, the feasibility of multi-person identification on low-cost commodity devices remains unclear. A critical question is whether weak multi-person performance is primarily an algorithmic limitation, or whether it reflects a more fundamental sensing ceiling on commodity WiFi hardware. We address this question through a systematic empirical study using commodity ESP32 WiFi sensors. We evaluated six different signal separation methods–FastICA, SOBI, PCA-ICA, NMF, Wavelet, and Tensor decomposition–across seven scenarios spanning 1-10 people in both controlled and realistic indoor environments. To investigate beyond classification accuracy, we introduce three diagnostic metrics: intra-subject variability (ISV), inter-subject distinguishability (ISD), and performance degradation rate (PDR). In all methods, performance remains moderate (39%-56% accuracy), with limited evidence that algorithmic choice alone solves the problem. The best-performing method, NMF, reaches 56% accuracy, while all methods exhibit extremely high feature-space overlap (97%-99%), unstable within-subject representations, and marked environmental sensitivity. These findings suggest that, under commodity ESP32 CSI constraints, dense multi-person gait identification is limited more by sensing quality and spatial diversity than by the chosen separation algorithm. Our results have direct implications for security and privacy: they call into question the practicality of commodity WiFi CSI as a robust multi-user biometric primitive for authentication, while also placing important bounds on the passive identification capabilities achievable with low-cost off-the-shelf WiFi hardware.

The ever-increasing adoption of Large Language Models in critical sectors like finance, healthcare, and government raises privacy concerns regarding the handling of sensitive Personally Identifiable Information (PII) during training. In response, regulations such as European Union's General Data Protection Regulation (GDPR) mandate the deletion of PII upon requests, underscoring the need for reliable and cost-effective data removal solutions. Machine unlearning has emerged as a promising direction for selectively forgetting data points. However, existing unlearning techniques typically apply a uniform forgetting strategy that neither accounts for the varying privacy risks posed by different PII attributes nor reflects associated business risks. In this work, we propose UnPII, the first PII-centric unlearning approach that prioritizes forgetting based on the risk of individual or combined PII attributes. To this end, we introduce the PII risk index (PRI), a composite metric that incorporates multiple dimensions of risk factors: identifiability, sensitivity, usability, linkability, permanency, exposability, and compliancy. The PRI enables a nuanced evaluation of privacy risks associated with PII exposures and can be tailored to align with organizational privacy policies. To support realistic assessment, we systematically construct a synthetic PII dataset (e.g., 1,700 PII instances) that simulates realistic exposure scenarios. UnPII seamlessly integrates with established unlearning algorithms, such as Gradient Ascent, Negative Preference Optimization, and Direct Preference Optimization, without modifying their underlying principles. Our experimental results demonstrate that UnPII achieves the improvements of accuracy up to 11.8%, utility up to 6.3%, and generalizability up to 12.4%, respectively, while incurring a modest fine-tuning overhead of 27.5% on average during unlearning.

Recent advancements in Large Vision-Language Models (LVLMs) have shown groundbreaking capabilities across diverse multimodal tasks. However, these models remain vulnerable to adversarial jailbreak attacks, where adversaries craft subtle perturbations to bypass safety mechanisms and trigger harmful outputs. Existing white-box attacks methods require full model accessibility, suffer from computing costs and exhibit insufficient adversarial transferability, making them impractical for real-world, black-box settings. To address these limitations, we propose a black-box jailbreak attack on LVLMs via Zeroth-Order optimization using Simultaneous Perturbation Stochastic Approximation (ZO-SPSA). ZO-SPSA provides three key advantages: (i) gradient-free approximation by input-output interactions without requiring model knowledge, (ii) model-agnostic optimization without the surrogate model and (iii) lower resource requirements with reduced GPU memory consumption. We evaluate ZO-SPSA on three LVLMs, including InstructBLIP, LLaVA and MiniGPT-4, achieving the highest jailbreak success rate of 83.0% on InstructBLIP, while maintaining imperceptible perturbations comparable to white-box methods. Moreover, adversarial examples generated from MiniGPT-4 exhibit strong transferability to other LVLMs, with ASR reaching 64.18%. These findings underscore the real-world feasibility of black-box jailbreaks and expose critical weaknesses in the safety mechanisms of current LVLMs

Federated Learning (FL) enables collaborative model training without direct data sharing, yet it remains vulnerable to privacy attacks such as model inversion and membership inference. Existing differential privacy (DP) solutions for FL often inject noise uniformly across the entire model, degrading utility while providing suboptimal privacy-utility tradeoffs. To address this, we propose LaDP, a novel layer-wise adaptive noise injection mechanism for FL that optimizes privacy protection while preserving model accuracy. LaDP leverages two key insights: (1) neural network layers contribute unevenly to model utility, and (2) layer-wise privacy leakage can be quantified via KL divergence between local and global model distributions. LaDP dynamically injects noise into selected layers based on their privacy sensitivity and importance to model performance. We provide a rigorous theoretical analysis, proving that LaDP satisfies $(\epsilon, \delta)$-DP guarantees and converges under bounded noise. Extensive experiments on CIFAR-10/100 datasets demonstrate that LaDP reduces noise injection by 46.14% on average compared to state-of-the-art (SOTA) methods while improving accuracy by 102.99%. Under the same privacy budget, LaDP outperforms SOTA solutions like Dynamic Privacy Allocation LDP and AdapLDP by 25.18% and 6.1% in accuracy, respectively. Additionally, LaDP robustly defends against reconstruction attacks, increasing the FID of the reconstructed private data by $>$12.84% compared to all baselines. Our work advances the practical deployment of privacy-preserving FL with minimal utility loss.

Autonomous AI agents executing multi-step tool sequences face semantic attacks that manifest in behavioral traces rather than isolated prompts. A critical challenge is cross-attack generalization: can detectors trained on known attack families recognize novel, unseen attack types? We discover that standard conversational tokenization – capturing linguistic patterns from agent interactions – fails catastrophically on structural attacks like tool hijacking (AUC 0.39) and data exfiltration (AUC 0.46), while succeeding on linguistic attacks like social engineering (AUC 0.78). We introduce structural tokenization, encoding execution-flow patterns (tool calls, arguments, observations) rather than conversational content. This simple representational change dramatically improves cross-attack generalization: +46 AUC points on tool hijacking, +39 points on data exfiltration, and +71 points on unknown attacks, while simultaneously improving in-distribution performance (+6 points). For attacks requiring linguistic features, we propose gated multi-view fusion that adaptively combines both representations, achieving AUC 0.89 on social engineering without sacrificing structural attack detection. Our findings reveal that AI agent security is fundamentally a structural problem: attack semantics reside in execution patterns, not surface language. While our rule-based tokenizer serves as a baseline, the structural abstraction principle generalizes even with simple implementation.

In recent years, cyberattacks - along with physical faults - have become an increasing factor causing system failures, especially in DER (Distributed Energy Resources) systems. In addition, according to the literature, a number of faults have been reported to remain undetected. Consequently, unlike anomaly detection works that only identify abnormalities, differentiating undetected faults and cyberattacks is a challenging task. Although several works have studied this problem, they crucially fall short of achieving an accurate distinction due to the reliance on physical laws or physical measurements. To resolve this issue, the industry typically conducts an integrated analysis with physical measurements and cyberspace information. Nevertheless, this industry approach consumes a significant amount of time due to the manual efforts required in the analysis. In this work, we focus on addressing these crucial gaps by proposing a non-trivial approach of distinguishing undetected faults and cyberattacks in DER systems. Specifically, first, a special kind of dependency graph is constructed using a novel virtual physical variable-oriented taint analysis (PVOTA) algorithm. Then, the graph is simplified using an innovative node pruning technique, which is based on a set of context-dependent operations. Next, a set of patterns capturing domain-specific knowledge is derived to bridge the semantic gaps between the cyber and physical sides. Finally, these patterns are matched to the relevant events that occurred during failure incidents, and possible root causes are concluded based on the pattern matching results. In the end, the efficacy of the proposed semi-automatic integrated analysis is evaluated through five case studies covering failure incidents caused by FDI attacks, system faults, and memory corruption attacks, including a false-classification scenario.

Jailbreak attacks present a significant challenge to the safety of Large Language Models (LLMs), yet current automated evaluation methods largely rely on coarse classifications that focus mainly on harmfulness, leading to substantial overestimation of attack success. To address this problem, we propose FJAR, a fine-grained jailbreak evaluation framework with anchored references. We first categorized jailbreak responses into five fine-grained categories: Rejective, Irrelevant, Unhelpful, Incorrect, and Successful, based on the degree to which the response addresses the malicious intent of the query. This categorization serves as the basis for FJAR. Then, we introduce a novel harmless tree decomposition approach to construct high-quality anchored references by breaking down the original queries. These references guide the evaluator in determining whether the response genuinely fulfills the original query. Extensive experiments demonstrate that FJAR achieves the highest alignment with human judgment and effectively identifies the root causes of jailbreak failures, providing actionable guidance for improving attack strategies.

Cybersecurity post-incident reviews are essential for identifying control failures and improving organisational resilience, yet they remain labour-intensive, time-consuming, and heavily reliant on expert judgment. This paper investigates whether Large Language Models (LLMs) can augment post-incident review workflows by autonomously analysing system evidence and identifying security policy gaps. We present a threat-informed, agentic framework that ingests log data, maps observed behaviours to the MITRE ATT&CK framework, and evaluates organisational security policies for adequacy and compliance. Using a simulated brute-force attack scenario against a Windows OpenSSH service (MITRE ATT&CK T1110), the system leverages GPT-4o for reasoning, LangGraph for multi-agent workflow orchestration, and LlamaIndex for traceable policy retrieval. Experimental results indicate that the LLM-based pipeline can interpret log-derived evidence, identify insufficient or missing policy controls, and generate actionable remediation recommendations with explicit evidence-to-policy traceability. Unlike prior work that treats log analysis and policy validation as isolated tasks, this study integrates both into a unified end-to-end proof-of-concept post-incident review framework. The findings suggest that LLM-assisted analysis has the potential to improve the efficiency, consistency, and auditability of post-incident evaluations, while highlighting the continued need for human oversight in high-stakes cybersecurity decision-making.

Private macOS frameworks underpin critical services and daemons but remain undocumented and distributed only as stripped binaries, complicating security analysis. We present MOTIF, an agentic framework that integrates tool-augmented analysis with a finetuned large language model specialized for Objective-C type inference. The agent manages runtime metadata extraction, binary inspection, and constraint checking, while the model generates candidate method signatures that are validated and refined into compilable headers. On MOTIF-Bench, a benchmark built from public frameworks with groundtruth headers, MOTIF improves signature recovery from 15% to 86% compared to baseline static analysis tooling, with consistent gains in tool-use correctness and inference stability. Case studies on private frameworks show that reconstructed headers compile, link, and facilitate downstream security research and vulnerability studies. By transforming opaque binaries into analyzable interfaces, MOTIF establishes a scalable foundation for systematic auditing of macOS internals.

This paper presents a comprehensive study of matrix Kloosterman sums, including their computational aspects, distributional behavior, and applications in cryptographic analysis. Building on the work of [Zelingher, 2023], we develop algorithms for evaluating these sums via Green's polynomials and establish a general framework for analyzing their statistical distributions. We further investigate the associated $L$-functions and clarify their relationships with symmetric functions and random matrix theory. We show that, analogous to the eigenvalue statistics of random matrices in compact Lie groups such as $SU(n)$ and $Sp(2n)$, the normalized values of matrix Kloosterman sums exhibit Sato-Tate equidistribution. Finally, we apply this framework to distinguish truly random sequences from those exhibiting subtle algebraic biases, and we propose a novel spectral test for cryptographic security based on the distributional signatures of matrix Kloosterman sums.

The rapid integration of Multimodal Large Language Models (MLLMs) into critical applications is increasingly hindered by persistent safety vulnerabilities. However, existing red-teaming benchmarks are often fragmented, limited to single-turn text interactions, and lack the scalability required for systematic evaluation. To address this, we introduce OpenRT, a unified, modular, and high-throughput red-teaming framework designed for comprehensive MLLM safety evaluation. At its core, OpenRT architects a paradigm shift in automated red-teaming by introducing an adversarial kernel that enables modular separation across five critical dimensions: model integration, dataset management, attack strategies, judging methods, and evaluation metrics. By standardizing attack interfaces, it decouples adversarial logic from a high-throughput asynchronous runtime, enabling systematic scaling across diverse models. Our framework integrates 37 diverse attack methodologies, spanning white-box gradients, multi-modal perturbations, and sophisticated multi-agent evolutionary strategies. Through an extensive empirical study on 20 advanced models (including GPT-5.2, Claude 4.5, and Gemini 3 Pro), we expose critical safety gaps: even frontier models fail to generalize across attack paradigms, with leading models exhibiting average Attack Success Rates as high as 49.14%. Notably, our findings reveal that reasoning models do not inherently possess superior robustness against complex, multi-turn jailbreaks. By open-sourcing OpenRT, we provide a sustainable, extensible, and continuously maintained infrastructure that accelerates the development and standardization of AI safety.

Integrated Sensing and Communication (ISAC) represents a significant shift in the 6G landscape, where wireless networks both sense the environment and communicate. While prior comprehensive surveys have established foundational elements of ISAC security, discussed perception-focused security models, and proposed layered defense strategies, this paper synthesizes these studies into a comprehensive taxonomic framework that covers the whole ISAC security domain. This paper provides a systematic and thorough review of ISAC security across multiple orthogonal dimensions. These include threat taxonomy and propagation methods; vulnerability analysis at design, physical, computational, and architectural levels; defense mechanisms categorized by deployment layer; security-performance trade-offs with theoretical bounds; sector-specific security demands for critical infrastructure; and emerging issues such as quantum resilience, AI-hardening, and privacy preservation. Unlike previous frameworks that primarily focus on vision, this review combines these dimensions, introduces new classification schemes that reveal hidden relationships between threats and defenses, and identifies key research gaps through structured analysis. This detailed taxonomy offers a valuable reference for researchers developing secure ISAC systems and policymakers establishing security standards.

The rigorous security model of Bitcoin's UTXO architecture often comes at the cost of developer usability, forcing a reliance on manual stack manipulation that leads to critical financial vulnerabilities like signature malleability, unspendable states and unconstrained execution paths. Industry standards such as Miniscript provide necessary abstractions for policy verification but do not model the full imperative logic required for complex contracts, leaving gaps in state management and resource liveness. This paper introduces Bithoven, a high-level language designed to bridge the gap between expressiveness and formal safety. By integrating a strict type checker and a resource liveness analyzer with a semantic control-flow analyzer, Bithoven eliminates major categories of consensus and logic defects defined in our fault model prior to deployment. Our results indicate that this safety comes at modest cost: Bithoven compiles to Bitcoin Script with efficiency comparable to hand-optimized code, demonstrating that type-safe, developer-friendly abstractions are viable even within the strict byte-size constraints of the Bitcoin blockchain.

The proliferation of Internet of Things (IoT) devices has introduced significant security challenges, primarily due to the opacity of firmware components and the complexity of supply chain dependencies. IoT firmware frequently relies on outdated, third-party libraries embedded within monolithic binary blobs, making vulnerability management difficult. While Software Bill of Materials (SBOM) standards have matured, generating actionable intelligence from raw firmware dumps remains a manual and error-prone process. This paper presents a lightweight, automated pipeline designed to extract file systems from Linux-based IoT firmware, generate a comprehensive SBOM, map identified components to known vulnerabilities, and apply a multi-factor triage scoring model. The proposed system focuses on risk prioritization by integrating signals from the Common Vulnerability Scoring System (CVSS), Exploit Prediction Scoring System (EPSS), and the CISA Known Exploited Vulnerabilities (KEV) catalog. Unlike conventional scanners that produce high volumes of uncontextualized alerts, this approach emphasizes triage by calculating a localized risk score for each finding. We describe the architecture, the normalization challenges of embedded Linux, and a scoring methodology intended to reduce alert fatigue. The study outlines a planned evaluation strategy to validate the extraction success rate and triage efficacy using a dataset of public vendor firmware, offering a reproducibility framework for future research in firmware security.

As frontier AIs become more powerful and costly to develop, adversaries have increasing incentives to steal model weights by mounting exfiltration attacks. In this work, we consider exfiltration attacks where an adversary attempts to sneak model weights out of a datacenter over a network. While exfiltration attacks are multi-step cyber attacks, we demonstrate that a single factor, the compressibility of model weights, significantly heightens exfiltration risk for large language models (LLMs). We tailor compression specifically for exfiltration by relaxing decompression constraints and demonstrate that attackers could achieve 16x to 100x compression with minimal trade-offs, reducing the time it would take for an attacker to illicitly transmit model weights from the defender's server from months to days. Finally, we study defenses designed to reduce exfiltration risk in three distinct ways: making models harder to compress, making them harder to 'find,' and tracking provenance for post-attack analysis using forensic watermarks. While all defenses are promising, the forensic watermark defense is both effective and cheap, and therefore is a particularly attractive lever for mitigating weight-exfiltration risk.

Data poisoning attacks (DPAs) are becoming popular as artificial intelligence (AI) algorithms, machine learning (ML) algorithms, and deep learning (DL) algorithms in this artificial intelligence (AI) era. Hackers and penetration testers are excessively injecting malicious contents in the training data (and in testing data too) that leads to false results that are very hard to inspect and predict. We have analyzed several recent technologies used (from deep reinforcement learning to federated learning) for the DPAs and their safety, security, & countermeasures. The problem setup along with the problem estimation is shown in the MuJoCo environment with performance of HalfCheetah before the dataset is poisoned and after the dataset is poisoned. We have analyzed several risks associated with the DPAs and falsification in medical data from popular poisoning data attacks to some popular data defenses. We have proposed robust offline reinforcement learning (Offline RL) for the safety and reliability with weighted hash verification along with density-ratio weighted behavioral cloning (DWBC) algorithm. The four stages of the proposed algorithm (as the Stage 0, the Stage 1, the Stage 2, and the Stage 3) are described with respect to offline RL, safety, and security for DPAs. The conclusion and future scope are provided with the intent to combine DWBC with other data defense strategies to counter and protect future contamination cyberattacks.

Trust and Reputation Management Systems (TRMSs) are critical for the modern web, yet their reliance on subjective user ratings or narrow Quality of Service (QoS) metrics lacks objective grounding. Concurrently, while regulatory frameworks like GDPR and HIPAA provide objective behavioral standards, automated compliance auditing has been limited to coarse, binary (pass/fail) outcomes. This paper bridges this research gap by operationalizing regulatory compliance as a quantitative and dynamic trust metric through our novel automated compliance engine (ACE). ACE first formalizes legal and organizational policies into a verifiable, obligation-centric logic. It then continuously audits system event logs against this logic to detect violations. The core of our contribution is a quantitative model that assesses the severity of each violation along multiple dimensions, including its Volume, Duration, Breadth, and Criticality, to compute a fine-grained, evolving compliance score. We evaluate ACE on a synthetic hospital dataset, demonstrating its ability to accurately detect a range of complex HIPAA and GDPR violations and produce a nuanced score that is significantly more expressive than traditional binary approaches. This work enables the development of more transparent, accountable, and resilient TRMSs on the Web.

Tool-augmented Large Language Model (LLM) agents create a new supply-chain surface: Model Context Protocol (MCP) tools are installed like third-party packages, yet their outputs can enter the agent's reasoning context. This enables confused-deputy risks in which attacker-controlled inputs cause otherwise benign tools to exercise legitimate authority over files, environment variables, or network-facing operations and reflect sensitive or instruction-like content into LLM-visible fields. We present SandScope, an MCP-aware audit framework that combines runtime witness detection with semantic tool profiling. SandScope executes portable tools under WebAssembly System Interface (WASI) or drives unmodified MCP servers over standard input/output (stdio), extracts LLM-visible sinks from tool results and prompt/message fields, and reports auditable source-to-sink witnesses from environment, file, and tool-input sources while separately recording network-intent and egress evidence. Its semantic layer recovers declared capabilities from tools/list metadata and static registrations to characterize attack surface when execution is incomplete. We evaluate SandScope on controlled cross-language subjects, an evasion benchmark, and a 100-repository MCP corpus. SandScope completes shallow dynamic scans for 35 repositories and, through a broader semantic profiling pass, recovers metadata for 1,127 tools across 71 repositories, including 886 tools with security-sensitive declared capabilities. A schema-guided exploration pass over the 35 dynamically scanned repositories re-executes 33 and observes source-to-sink witnesses in 12. These results show that SandScope provides practical, auditable evidence for MCP tool risk through controlled execution, MCP-aware sink extraction, runtime witness reporting, and semantic capability profiling.

The rapid advancements in artificial intelligence have significantly accelerated the adoption of speech recognition technology, leading to its widespread integration across various applications. However, this surge in usage also highlights a critical issue: audio data is highly vulnerable to unauthorized exposure and analysis, posing significant privacy risks for businesses and individuals. This paper introduces an Information-Obfuscation Reversible Adversarial Example (IO-RAE) framework, the pioneering method designed to safeguard audio privacy using reversible adversarial examples. IO-RAE leverages large language models to generate misleading yet contextually coherent content, effectively preventing unauthorized eavesdropping by humans and Automatic Speech Recognition (ASR) systems. Additionally, we propose the Cumulative Signal Attack technique, which mitigates high-frequency noise and enhances attack efficacy by targeting low-frequency signals. Our approach ensures the protection of audio data without degrading its quality or our ability. Experimental evaluations demonstrate the superiority of our method, achieving a targeted misguidance rate of 96.5% and a remarkable 100% untargeted misguidance rate in obfuscating target keywords across multiple ASR models, including a commercial black-box system from Google. Furthermore, the quality of the recovered audio, measured by the Perceptual Evaluation of Speech Quality score, reached 4.45, comparable to high-quality original recordings. Notably, the recovered audio processed by ASR systems exhibited an error rate of 0%, indicating nearly lossless recovery. These results highlight the practical applicability and effectiveness of our IO-RAE framework in protecting sensitive audio privacy.

Confidential containers protect cloud-native workloads using trusted execution environments (TEEs). However, existing Container-in-TEE designs (e.g., Confidential Containers (CoCo)) encapsulate the entire runtime within the TEE, inflating the trusted computing base (TCB) and introducing redundant components and cross-layer overhead. We present Arca, a lightweight confidential container framework based on a TEE-in-Container architecture that isolates each workload in an independent, hardware-enforced trust domain while keeping orchestration logic outside the TEE. This design minimizes inter-layer dependencies, confines compromise to per-container boundaries, and restores the TEE's minimal trust principle. We implemented Arca on Intel SGX, Intel TDX, and AMD SEV. Experimental results show that Arca achieves near-native performance and outperforms CoCo in most benchmarks, while the reduced TCB significantly improves verifiability and resilience against host-level compromise. Arca emonstrates that efficient container management and strong runtime confidentiality can be achieved without sacrificing security assurance.

Large Language Models (LLMs) can generate plausible code, but in settings that require exact stdin/stdout behavior they frequently produce programs that compile yet fail tests, and in some cases they introduce security-sensitive patterns. This paper presents SecureCodeRL, a reinforcement learning (RL) pipeline for security-aware code generation that optimizes a combined reward R = {\alpha}Rfunc + e̱ta}Rsec. The key idea is a partial-credit functional reward that assigns intermediate scores for syntactic validity, successful execution, and producing output, reducing reward sparsity that otherwise stalls learning on competitive programming style tasks. I evaluate supervised fine-tuning (SFT) and PPO variants on a small held-out prompt set from APPS+ and observe that PPO with partial credit (using a continued-training variant) improves syntax validity from 45% (SFT) to 60% and achieves the only non-zero test success signal in this pilot evaluation (5% at-least-one-test-pass), while remaining 100% clean under Bandit static analysis. Although Bandit findings were absent in this small evaluation, the security term is integrated into training to discourage insecure shortcuts when they appear.

Encrypted network traffic poses significant challenges for intrusion detection due to the lack of payload visibility, limited labeled datasets, and high class imbalance between benign and malicious activities. Traditional data augmentation methods struggle to preserve the complex temporal and statistical characteristics of real network traffic. To address these issues, this work explores the use of Generative AI (GAI) models to synthesize realistic and diverse encrypted traffic traces. We evaluate three approaches: Variational Autoencoders (VAE), Generative Adversarial Networks (GAN), and SMOTE (Synthetic Minority Over-sampling Technique), each integrated with a preprocessing pipeline that includes feature selection and class balancing. The UNSW NB-15 dataset is used as the primary benchmark, focusing on Tor traffic as anomalies. We analyze statistical similarity between real and synthetic data, and assess classifier performance using metrics such as Accuracy, F1-score, and AUC-ROC. Results show that VAE-generated data provides the best balance between privacy and performance, while GANs offer higher fidelity but risk overfitting. SMOTE, though simple, enhances recall but may lack diversity. The findings demonstrate that GAI methods can significantly improve encrypted traffic detection when trained with privacy-preserving synthetic data.

Cybersecurity poses considerable problems to Cloud Computing (CC), especially regarding Intrusion Detection Systems (IDSs), facing difficulties with skewed datasets and suboptimal classification model performance. This study presents the Hybrid Intrusion Detection System (HyIDS), an innovative IDS that employs the Energy Valley Optimizer (EVO) for Feature Selection (FS). Additionally, it introduces a novel technique for enhancing the cybersecurity of cloud computing through the integration of machine learning methodologies with the EVO Algorithm. The Energy Valley Optimizer (EVO) effectively diminished features in the CIC-DDoS2019 dataset from 88 to 38 and in the CSE-CIC-IDS2018 data from 80 to 43, significantly enhancing computing efficiency. HyIDS incorporates four Machine Learning (ML) models: Support Vector Machine (SVM), Random Forest (RF), Decision Tree (D_Tree), and K-Nearest Neighbors (KNN). The proposed HyIDS was assessed utilizing two real-world intrusion datasets, CIC-DDoS2019 and CSE-CIC-IDS2018, both distinguished by considerable class imbalances. The CIC-DDoS2019 dataset has a significant imbalance between DDoS assault samples and legal traffic, while the CSE-CIC-IDS2018 dataset primarily comprises benign traffic with insufficient representation of attack types, complicating the detection of minority attacks. A downsampling technique was employed to balance the datasets, hence improving detection efficacy for both benign and malicious traffic. Twenty-four trials were done, revealing substantial enhancements in categorization accuracy, precision, and recall. Our suggested D_TreeEVO model attained an accuracy rate of 99.13% and an F1 score of 98.94% on the CIC-DDoS2019 dataset, and an accuracy rate of 99.78% and an F1 score of 99.70% on the CSE-CIC-IDS2018 data. These data demonstrate that EVO significantly improves cybersecurity in Cloud Computing (CC).

The strategy of combining diffusion-based generative models with classifiers continues to demonstrate state-of-the-art performance on adversarial robustness benchmarks. Known as adversarial purification, this exploits a diffusion model's capability of identifying high density regions in data distributions to purify adversarial perturbations from inputs. However, existing diffusion-based purification defenses are impractically slow and limited in robustness due to the low levels of noise used in the diffusion process. This low noise design aims to preserve the semantic features of the original input, thereby minimizing utility loss for benign inputs. Our findings indicate that systematic amplification of noise throughout the diffusion process improves the robustness of adversarial purification. However, this approach presents a key challenge, as noise levels cannot be arbitrarily increased without risking distortion of the input. To address this key problem, we introduce high levels of noise during the forward process and propose the ring proximity correction to gradually eliminate adversarial perturbations whilst closely preserving the original data sample. As a second contribution, we propose a new stochastic sampling method which introduces additional noise during the reverse diffusion process to dilute adversarial perturbations. Without relying on gradient obfuscation, these contributions result in a new robustness accuracy record of 44.23% on ImageNet using AutoAttack ($\ell_{\infty}=4/255$), an improvement of +2.07% over the previous best work. Furthermore, our method reduces inference time to 1.08 seconds per sample on ImageNet, a $47\times$ improvement over the existing state-of-the-art approach, making it far more practical for real-world defensive scenarios.

As quantum computing advances, the cryptographic algorithms that underpin confidentiality, integrity, and authentication in Intelligent Transportation Systems (ITS) face increasing vulnerability to quantum-enabled attacks. To address these risks, governments and industry stakeholders are turning toward post-quantum cryptography (PQC), a class of algorithms designed to resist adversaries equipped with quantum computing capabilities. However, existing studies provide limited insight into the implementation-focused aspects of PQC in the ITS domain. This review addresses that gap by evaluating the readiness of vehicular communication and security standards for adopting PQC. It examines in-vehicle networks and vehicle-to-everything (V2X) interfaces, and investigates vulnerabilities at the physical implementation layer of cryptographic hardware and embedded platforms, primarily exposure to side-channel and fault injection attacks. The review identifies thirteen research gaps: non-PQC-ready standards; constraints in embedded implementation and hybrid cryptography; interoperability and certificate-management barriers; a lack of real-world PQC deployment data in ITS; and physical-attack vulnerabilities in PQC-enabled vehicular communication. We present several future research directions, including updating vehicular communication and security standards, optimizing PQC for low-power devices, enhancing interoperability and certificate-management frameworks for PQC integration, conducting real-world evaluations of PQC-enabled communication and control functions across ITS deployments, and strengthening defenses against AI-assisted physical attacks. A phased roadmap is presented that aligns PQC deployment with regulatory, performance, and safety requirements, thereby guiding the secure evolution of ITS in the quantum computing era.

We show that a simple eavesdropper listening in on classical communication between potentially entangled quantum parties will eventually be able to impersonate any of the parties. Furthermore, the attack is efficient if one-way puzzles do not exist. As a direct consequence, one-way puzzles are implied by reusable authentication schemes over classical channels with quantum pre-shared secrets that are potentially evolving. As an additional application, we show that any quantum money scheme that can be verified through only classical queries to any oracle cannot be information-theoretically secure. This significantly generalizes the prior work by Ananth, Hu, and Yuen (ASIACRYPT'23) where they showed the same but only for the specific case of random oracles. Therefore, verifying black-box constructions of quantum money inherently requires coherently evaluating the underlying cryptographic tools, which may be difficult for near-term quantum devices.

Out-of-band screening of microcontrollers is a major gap in semiconductor supply chain security. High-assurance techniques such as X-ray and destructive reverse engineering are accurate but slow and expensive, hindering comprehensive detection for hardware Trojans or firmware tampering. Consequently, there has been increased interest in applying machine learning techniques to automate forensic examination, enabling rapid, large-scale inspection of components without manual oversight. We introduce a non-destructive screening method that uses power side-channel measurements and generative modeling to detect tampering in commodity microcontrollers without trusted hardware. As a proof-of-concept, differential power analysis (DPA) traces are collected from the ChipWhisperer and a generative adversarial network (GAN) is trained only on benign measurements to learn nominal power behavior. The trained discriminator then serves as a one-class anomaly detector. We report detection performance on multiple tampering scenarios and discuss how this technique can serve as an intermediate screening tier between basic functional tests and high-cost forensic analysis. The proposed method is evaluated in the context of semiconductor supply chain practice and policy to assess its suitability as an intermediate assurance mechanism.

The proliferation of Internet of Things devices in critical infrastructure has created unprecedented cybersecurity challenges, necessitating collaborative threat detection mechanisms that preserve data privacy while maintaining robustness against sophisticated attacks. Traditional federated learning approaches for IoT security suffer from two critical vulnerabilities: susceptibility to Byzantine attacks where malicious participants poison model updates, and inadequacy against future quantum computing threats that can compromise cryptographic aggregation protocols. This paper presents a novel Byzantine-robust federated learning framework integrated with post-quantum secure aggregation specifically designed for real-time threat intelligence sharing across critical IoT infrastructure. The proposed framework combines a adaptive weighted aggregation mechanism with lattice-based cryptographic protocols to simultaneously defend against model poisoning attacks and quantum adversaries. We introduce a reputation-based client selection algorithm that dynamically identifies and excludes Byzantine participants while maintaining differential privacy guarantees. The secure aggregation protocol employs CRYSTALS-Kyber for key encapsulation and homomorphic encryption to ensure confidentiality during parameter updates. Experimental evaluation on industrial IoT intrusion detection datasets demonstrates that our framework achieves 96.8% threat detection accuracy while successfully mitigating up to 40% Byzantine attackers, with only 18% computational overhead compared to non-secure federated approaches. The framework maintains sub-second aggregation latency suitable for real-time applications and provides 256-bit post-quantum security level.

GPUs have gained significant popularity over the past decade, extending beyond their original role in graphics rendering. This evolution has brought GPU security and reliability to the forefront of concerns. Prior research has shown that CUDA's lack of memory safety can lead to serious vulnerabilities. While fuzzing is effective for finding such bugs on CPUs, equivalent tools for GPUs are lacking due to architectural differences and lack of built-in error detection. In this paper, we propose CuFuzz, a novel compiler-runtime co-design solution to extend state-of-the-art CPU fuzzing tools to GPU programs. CuFuzz transforms GPU programs into CPU programs using compiler IR-level transformations to enable effective fuzz testing. To the best of our knowledge, CuFuzz is the first mechanism to bring fuzzing support to CUDA, addressing a critical gap in GPU security research. By leveraging CPU memory error detectors such as Address Sanitizer, CuFuzz aims to uncover memory safety bugs and related correctness vulnerabilities in CUDA code, enhancing the security and reliability of GPU-accelerated applications. To ensure high fuzzing throughput, we introduce two compiler-runtime co-optimizations tailored for GPU code: Partial Representative Execution (PREX) and Access-Index Preserving Pruning (AXIPrune), achieving average throughput improvements of 32x with PREX and an additional 33% gain with AXIPrune on top of PREX-optimized code. Together, these optimizations can yield up to a 224.31x speedup. In our fuzzing campaigns, CuFuzz uncovered 122 security vulnerabilities in widely used benchmarks.