arXiv:2603.28798v1 Announce Type: cross Abstract: Physically Unclonable Functions (PUFs) provide promising hardware security for IoT authentication, leveraging inherent randomness suitable for resource constrained environments. However, ML/DL modeling attacks threaten PUF security by learning challenge-response patterns. This work introduces a custom resistor-capacitor (RC) based dynamically reconfigurable PUF using 32-bit challenge-response pairs (CRPs) designed to resist such attacks. We systematically evaluated robustness by generating a CRP dataset and splitting it into training, validation, and test sets. Multiple ML techniques including Artificial Neural Networks (ANN), Gradient Boosted Neural Networks (GBNN), Decision Trees (DT), Random Forests (RF), and XGBoost, were trained to model PUF behavior. While all models achieved 100% training accuracy, test performance remained near random guessing: 51.05% (ANN), 53.27% (GBNN), 50.06% (DT), 52.08% (RF), and 50.97% (XGBoost). These results demonstrate the proposed PUF's strong resistance to ML-driven modeling attacks, as advanced algorithms fail to reproduce accurate responses. The dynamically reconfigurable architecture enhances robustness against adversarial threats with minimal resource overhead. This simple RC-PUF offers an effective, low-cost alternative to complex encryption for securing next-generation IoT authentication against machine learning-based threats, ensuring reliable device verification without compromising computational efficiency or scalability in deployed IoT networks.

arXiv:2603.28815v1 Announce Type: cross Abstract: This technical report presents SkillTester, a tool for evaluating the utility and security of agent skills. Its evaluation framework combines paired baseline and with-skill execution conditions with a separate security probe suite. Grounded in a comparative utility principle and a user-facing simplicity principle, the framework normalizes raw execution artifacts into a utility score, a security score, and a three-level security status label. More broadly, it can be understood as a comparative quality-assurance harness for agent skills in an agent-first world. The public service is deployed at https://skilltester.ai, and the broader project is maintained at https://github.com/skilltester-ai/skilltester.

arXiv:2603.28817v1 Announce Type: cross Abstract: Small Language Models (SLMs) are emerging as efficient and economically viable alternatives to Large Language Models (LLMs), offering competitive performance with significantly lower computational costs and latency. These advantages make SLMs suitable for resource-constrained and efficient deployment on edge devices. However, existing jailbreak defenses show limited robustness against heterogeneous attacks, largely due to an incomplete understanding of the internal representations across different layers of language models that facilitate jailbreak behaviors. In this paper, we conduct a comprehensive empirical study on 9 jailbreak attacks across 7 SLMs and 3 LLMs. Our analysis shows that SLMs remain highly vulnerable to malicious prompts that bypass safety alignment. We analyze hidden-layer activations across different layers and model architectures, revealing that different input types form distinguishable patterns in the internal representation space. Based on this observation, we propose GUARD-SLM, a lightweight token activation-based method that operates in the representation space to filter malicious prompts during inference while preserving benign ones. Our findings highlight robustness limitations across layers of language models and provide a practical direction for secure small language model deployment.

arXiv:2603.28824v1 Announce Type: cross Abstract: Dataset condensation aims to synthesize compact yet informative datasets that retain the training efficacy of full-scale data, offering substantial gains in efficiency. Recent studies reveal that the condensation process can be vulnerable to backdoor attacks, where malicious triggers are injected into the condensation dataset, manipulating model behavior during inference. While prior approaches have made progress in balancing attack success rate and clean test accuracy, they often fall short in preserving stealthiness, especially in concealing the visual artifacts of condensed data or the perturbations introduced during inference. To address this challenge, we introduce Sneakdoor, which enhances stealthiness without compromising attack effectiveness. Sneakdoor exploits the inherent vulnerability of class decision boundaries and incorporates a generative module that constructs input-aware triggers aligned with local feature geometry, thereby minimizing detectability. This joint design enables the attack to remain imperceptible to both human inspection and statistical detection. Extensive experiments across multiple datasets demonstrate that Sneakdoor achieves a compelling balance among attack success rate, clean test accuracy, and stealthiness, substantially improving the invisibility of both the synthetic data and triggered samples while maintaining high attack efficacy. The code is available at https://github.com/XJTU-AI-Lab/SneakDoor.

arXiv:2603.28838v1 Announce Type: cross Abstract: Intrusion Detection System (IDS) is often calibrated to known attacks and generalizes poorly to unknown threats. This paper proposes GMA-SAWGAN-GP, a novel generative augmentation framework built on a Self-Attention-enhanced Wasserstein GAN with Gradient Penalty (WGAN-GP). The generator employs Gumbel-Softmax regularization to model discrete fields, while a Multilayer Perceptron (MLP)-based AutoEncoder acts as a manifold regularizer. A lightweight gating network adaptively balances adversarial and reconstruction losses via entropy regularization, improving stability and mitigating mode collapse. The self-attention mechanism enables the generator to capture both short- and long-range dependencies among features within each record while preserving categorical semantics through Gumbel-Softmax heads. Extensive experiments on NSL-KDD, UNSW-NB15, and CICIDS2017 using five representative IDS models demonstrate that GMA-SAWGAN-GP significantly improves detection performance on known attacks and enhances generalization to unknown attacks. Leave-One-Attack-type-Out (LOAO) evaluations using Area Under the Receiver Operating Characteristic (AUROC) and True Positive Rate at a 5 percent False Positive Rate confirm that IDS models trained on augmented datasets achieve higher robustness under unseen attack scenarios. Ablation studies validate the contribution of each component to performance gains. Compared with baseline models, the proposed framework improves binary classification accuracy by an average of 5.3 percent and multi-classification accuracy by 2.2 percent, while AUROC and True Positive Rate at a 5 percent False Positive Rate for unknown attacks increase by 3.9 percent and 4.8 percent, respectively, across the three datasets. Overall, GMA-SAWGAN-GP provides an effective approach to generative augmentation for mixed-type network traffic, improving IDS accuracy and resilience.

arXiv:2603.28972v1 Announce Type: cross Abstract: The large-scale adoption of Large Language Models (LLMs) forces a trade-off between operational cost (OpEx) and data privacy. Current routing frameworks reduce costs but ignore prompt sensitivity, exposing users and institutions to leakage risks towards third-party cloud providers. We formalise the "Inseparability Paradigm": advanced context management intrinsically coincides with privacy management. We propose a local "Privacy Guard" -- a holistic contextual observer powered by an on-premise Small Language Model (SLM) -- that performs abstractive summarisation and Automatic Prompt Optimisation (APO) to decompose prompts into focused sub-tasks, re-routing high-risk queries to Zero-Trust or NDA-covered models. This dual mechanism simultaneously eliminates sensitive inference vectors (Zero Leakage) and reduces cloud token payloads (OpEx Reduction). A LIFO-based context compacting mechanism further bounds working memory, limiting the emergent leakage surface. We validate the framework through a 2x2 benchmark (Lazy vs. Expert users; Personal vs. Institutional secrets) on a 1,000-sample dataset, achieving a 45% blended OpEx reduction, 100% redaction success on personal secrets, and -- via LLM-as-a-Judge evaluation -- an 85% preference rate for APO-compressed responses over raw baselines. Our results demonstrate that Token Parsimony and Zero Leakage are mathematically dual projections of the same contextual compression operator.

arXiv:2603.28998v1 Announce Type: cross Abstract: As Large Language Models (LLMs) and multi-agent AI systems are demonstrating increasing potential in cybersecurity operations, organizations, policymakers, model providers, and researchers in the AI and cybersecurity communities are interested in quantifying the capabilities of such AI systems to achieve more autonomous SOCs (security operation centers) and reduce manual effort. In particular, the AI and cybersecurity communities have recently developed several benchmarks for evaluating the red team capabilities of multi-agent AI systems. However, because the operations in SOCs are dominated by blue team operations, the capabilities of AI systems & agents to achieve more autonomous SOCs cannot be evaluated without a benchmark focused on blue team operations. To our best knowledge, no systematic benchmark for evaluating coordinated multi-task blue team AI has been proposed in the literature. Existing blue team benchmarks focus on a particular task. The goal of this work is to develop a set of design principles for the construction of a benchmark, which is denoted as SOC-bench, to evaluate the blue team capabilities of AI. Following these design principles, we have developed a conceptual design of SOC-bench, which consists of a family of five blue team tasks in the context of large-scale ransomware attack incident response.

arXiv:2603.29038v1 Announce Type: cross Abstract: Fine-tuning APIs offered by major AI providers create new attack surfaces where adversaries can bypass safety measures through targeted fine-tuning. We introduce Trojan-Speak, an adversarial fine-tuning method that bypasses Anthropic's Constitutional Classifiers. Our approach uses curriculum learning combined with GRPO-based hybrid reinforcement learning to teach models a communication protocol that evades LLM-based content classification. Crucially, while prior adversarial fine-tuning approaches report more than 25% capability degradation on reasoning benchmarks, Trojan-Speak incurs less than 5% degradation while achieving 99+% classifier evasion for models with 14B+ parameters. We demonstrate that fine-tuned models can provide detailed responses to expert-level CBRN (Chemical, Biological, Radiological, and Nuclear) queries from Anthropic's Constitutional Classifiers bug-bounty program. Our findings reveal that LLM-based content classifiers alone are insufficient for preventing dangerous information disclosure when adversaries have fine-tuning access, and we show that activation-level probes can substantially improve robustness to such attacks.

arXiv:2603.29062v1 Announce Type: cross Abstract: LLM-based chatbots in government services face critical security gaps. Multi-turn adversarial attacks achieve over 90% success against current defenses, and single-layer guardrails are bypassed with similar rates. We present CivicShield, a cross-domain defense-in-depth framework for government-facing AI chatbots. Drawing on network security, formal verification, biological immune systems, aviation safety, and zero-trust cryptography, CivicShield introduces seven defense layers: (1) zero-trust foundation with capability-based access control, (2) perimeter input validation, (3) semantic firewall with intent classification, (4) conversation state machine with safety invariants, (5) behavioral anomaly detection, (6) multi-model consensus verification, and (7) graduated human-in-the-loop escalation. We present a formal threat model covering 8 multi-turn attack families, map the framework to NIST SP 800-53 controls across 14 families, and evaluate using ablation analysis. Theoretical analysis shows layered defenses reduce attack probability by 1-2 orders of magnitude versus single-layer approaches. Simulation against 1,436 scenarios including HarmBench (416), JailbreakBench (200), and XSTest (450) achieves 72.9% combined detection [69.5-76.0% CI] with 2.9% effective false positive rate after graduated response, while maintaining 100% detection of multi-turn crescendo and slow-drift attacks. The honest drop on real benchmarks versus author-generated scenarios (71.2% vs 76.7% on HarmBench, 47.0% vs 70.0% on JailbreakBench) validates independent evaluation importance. CivicShield addresses an open gap at the intersection of AI safety, government compliance, and practical deployment.

arXiv:2603.29216v1 Announce Type: cross Abstract: Large Language Models (LLMs) have emerged as a popular choice in vulnerability detection studies given their foundational capabilities, open source availability, and variety of models, but have limited scalability due to extensive compute requirements. Using the natural graph relational structure of code, we show that our proposed graph neural network (GNN) based deep learning model VulGNN for vulnerability detection can achieve performance almost on par with LLMs, but is 100 times smaller in size and fast to retrain and customize. We describe the VulGNN architecture, ablation studies on components, learning rates, and generalizability to different code datasets. As a lightweight model for vulnerability analysis, VulGNN is efficient and deployable at the edge as part of real-world software development pipelines.

arXiv:2603.29289v1 Announce Type: cross Abstract: The fast pace of modern AI is rapidly transforming traditional industrial systems into vast, intelligent and potentially unmanned autonomous operational environments driven by AI-based solutions. These solutions leverage various forms of machine learning, reinforcement learning, and generative AI. The introduction of such smart capabilities has pushed the envelope in multiple industrial domains, enabling predictive maintenance, optimized performance, and streamlined workflows. These solutions are often deployed across the Industrial Internet of Things (IIoT) and supported by the Edge-Fog-Cloud computing continuum to enable urgent (i.e., real-time or near real-time) decision-making. Despite the current trend of aggressively adopting these smart industrial solutions to increase profit, quality, and efficiency, large-scale integration and deployment also bring serious hazards that if ignored can undermine the benefits of smart industries. These hazards include unforeseen interoperability side-effects and heightened vulnerability to cyber threats, particularly in environments operating with a plethora of heterogeneous IIoT systems. The goal of this study is to shed light on the potential consequences of industrial smartness, with a particular focus on security implications, including vulnerabilities, side effects, and cyber threats. We distinguish software-level downsides stemming from both traditional AI solutions and generative AI from those originating in the infrastructure layer, namely IIoT and the Edge-Cloud continuum. At each level, we investigate potential vulnerabilities, cyber threats, and unintended side effects. As industries continue to become smarter, understanding and addressing these downsides will be crucial to ensure secure and sustainable development of smart industrial systems.

arXiv:2603.29328v1 Announce Type: cross Abstract: Backdoor attacks on federated learning (FL) are most often evaluated with synthetic corner patches or out-of-distribution (OOD) patterns that are unlikely to arise in practice. In this paper, we revisit the backdoor threat to standard FL (a single global model) under a more realistic setting where triggers must be semantically meaningful, in-distribution, and visually plausible. We propose SABLE, a Semantics-Aware Backdoor for LEarning in federated settings, which constructs natural, content-consistent triggers (e.g., semantic attribute changes such as sunglasses) and optimizes an aggregation-aware malicious objective with feature separation and parameter regularization to keep attacker updates close to benign ones. We instantiate SABLE on CelebA hair-color classification and the German Traffic Sign Recognition Benchmark (GTSRB), poisoning only a small, interpretable subset of each malicious client's local data while otherwise following the standard FL protocol. Across heterogeneous client partitions and multiple aggregation rules (FedAvg, Trimmed Mean, MultiKrum, and FLAME), our semantics-driven triggers achieve high targeted attack success rates while preserving benign test accuracy. These results show that semantics-aligned backdoors remain a potent and practical threat in federated learning, and that robustness claims based solely on synthetic patch triggers can be overly optimistic.

arXiv:2603.29403v1 Announce Type: cross Abstract: LLM-as-a-Judge (LaaJ) is a novel paradigm in which powerful language models are used to assess the quality, safety, or correctness of generated outputs. While this paradigm has significantly improved the scalability and efficiency of evaluation processes, it also introduces novel security risks and reliability concerns that remain largely unexplored. In particular, LLM-based judges can become both targets of adversarial manipulation and instruments through which attacks are conducted, potentially compromising the trustworthiness of evaluation pipelines. In this paper, we present the first Systematization of Knowledge (SoK) focusing on the security aspects of LLM-as-a-Judge systems. We perform a comprehensive literature review across major academic databases, analyzing 863 works and selecting 45 relevant studies published between 2020 and 2026. Based on this study, we propose a taxonomy that organizes recent research according to the role played by LLM-as-a-Judge in the security landscape, distinguishing between attacks targeting LaaJ systems, attacks performed through LaaJ, defenses leveraging LaaJ for security purposes, and applications where LaaJ is used as an evaluation strategy in security-related domains. We further provide a comparative analysis of existing approaches, highlighting current limitations, emerging threats, and open research challenges. Our findings reveal significant vulnerabilities in LLM-based evaluation frameworks, as well as promising directions for improving their robustness and reliability. Finally, we outline key research opportunities that can guide the development of more secure and trustworthy LLM-as-a-Judge systems.

arXiv:2603.29520v1 Announce Type: cross Abstract: Encrypted traffic classification is a critical task for network security. While deep learning has advanced this field, the occlusion of payload semantics by encryption severely challenges standard modeling approaches. Most existing frameworks rely on static and homogeneous pipelines that apply uniform parameter sharing and static fusion strategies across all inputs. This one-size-fits-all static design is inherently flawed: by forcing structured headers and randomized payloads into a unified processing pipeline, it inevitably entangles the raw protocol signals with stochastic encryption noise, thereby degrading the fine-grained discriminative features. In this paper, we propose TrafficMoE, a framework that breaks through the bottleneck of static modeling by establishing a Disentangle-Filter-Aggregate (DFA) paradigm. Specifically, to resolve the structural between-components conflict, the architecture disentangles headers and payloads using dual-branch sparse Mixture-of-Experts (MoE), enabling modality-specific modeling. To mitigate the impact of stochastic noise, an uncertainty-aware filtering mechanism is introduced to quantify reliability and selectively suppress high-variance representations. Finally, to overcome the limitations of static fusion, a routing-guided strategy aggregates cross-modality features dynamically, that adaptively weighs contributions based on traffic context. With this DFA paradigm, TrafficMoE maximizes representational efficiency by focusing solely on the most discriminative traffic features. Extensive experiments on six datasets demonstrate TrafficMoE consistently outperforms state-of-the-art methods, validating the necessity of heterogeneity-aware modeling in encrypted traffic analysis. The source code is publicly available at https://github.com/Posuly/TrafficMoE_main.

arXiv:2603.29537v1 Announce Type: cross Abstract: Network traffic classification using self-supervised pre-training models based on Masked Autoencoders (MAE) has demonstrated a huge potential. However, existing methods are confined to isolated byte-level reconstruction of individual flows, lacking adequate perception of the multi-granularity contextual relationship in traffic. To address this limitation, we propose Mean MAE (MMAE), a teacher-student MAE paradigm with flow mixing strategy for building encrypted traffic pre-training model. MMAE employs a self-distillation mechanism for teacher-student interaction, where the teacher provides unmasked flow-level semantic supervision to advance the student from local byte reconstruction to multi-granularity comprehension. To break the information bottleneck in individual flows, we introduce a dynamic Flow Mixing (FlowMix) strategy to replace traditional random masking mechanism. By constructing challenging cross-flow mixed samples with interferences, it compels the model to learn discriminative representations from distorted tokens. Furthermore, we design a Packet-importance aware Mask Predictor (PMP) equipped with an attention bias mechanism that leverages packet-level side-channel statistics to dynamically mask tokens with high semantic density. Numerous experiments on a number of datasets covering encrypted applications, malware, and attack traffic demonstrate that MMAE achieves state-of-the-art performance. The code is available at https://github.com/lx6c78/MMAE

arXiv:2603.30016v1 Announce Type: cross Abstract: AI agents, predominantly powered by large language models (LLMs), are vulnerable to indirect prompt injection, in which malicious instructions embedded in untrusted data can trigger dangerous agent actions. This position paper discusses our vision for system-level defenses against indirect prompt injection attacks. We articulate three positions: (1) dynamic replanning and security policy updates are often necessary for dynamic tasks and realistic environments; (2) certain context-dependent security decisions would still require LLMs (or other learned models), but should only be made within system designs that strictly constrain what the model can observe and decide; (3) in inherently ambiguous cases, personalization and human interaction should be treated as core design considerations. In addition to our main positions, we discuss limitations of existing benchmarks that can create a false sense of utility and security. We also highlight the value of system-level defenses, which serve as the skeleton of agentic systems by structuring and controlling agent behaviors, integrating rule-based and model-based security checks, and enabling more targeted research on model robustness and human interaction.

arXiv:2507.10610v2 Announce Type: replace-cross Abstract: Graphical user interface (GUI) agents built on multimodal large language models (MLLMs) have recently demonstrated strong decision-making abilities in screen-based interaction tasks. However, they remain highly vulnerable to pop-up-based environmental injection attacks, where malicious visual elements divert model attention and lead to unsafe or incorrect actions. Existing defense methods either require costly retraining or perform poorly under inductive interference. In this work, we systematically study how such attacks alter the attention behavior of GUI agents and uncover a layer-wise attention divergence pattern between correct and incorrect outputs. Based on this insight, we propose \textbf{LaSM}, a \textit{Layer-wise Scaling Mechanism} that selectively amplifies attention and MLP modules in critical layers. LaSM improves the alignment between model saliency and task-relevant regions without additional training. Extensive experiments across multiple datasets demonstrate that our method significantly improves the defense success rate and exhibits strong robustness, while having negligible impact on the model's general capabilities. Our findings reveal that attention misalignment is a core vulnerability in MLLM agents and can be effectively addressed through selective layer-wise modulation. Our code can be found in https://github.com/YANGTUOMAO/LaSM.

arXiv:2509.22097v2 Announce Type: replace-cross Abstract: Large language model-powered code agents are rapidly transforming software engineering, yet the security risks of their generated code have become a critical concern. Existing benchmarks have provided valuable insights, but they fail to capture scenarios in which vulnerabilities are actually introduced by human developers, making fair comparisons between humans and agents infeasible. We therefore introduce SecureVibeBench, a benchmark of 105 C/C++ secure coding tasks sourced from 41 projects in OSS-Fuzz for code agents. SecureVibeBench has the following features: (i) realistic task settings that require multi-file edits in large repositories, (ii)~aligned contexts based on real-world open-source vulnerabilities with precisely identified vulnerability introduction points, and (iii) comprehensive evaluation that combines functionality testing and security checking with both static and dynamic oracles. We evaluate 5 popular code agents like OpenHands, supported by 5 LLMs (e.g., Claude sonnet 4.5) on SecureVibeBench. Results show that current agents struggle to produce both correct and secure code, as even the best-performing one, produces merely 23.8\% correct and secure solutions on SecureVibeBench.

arXiv:2602.10149v3 Announce Type: replace-cross Abstract: Third-Party Risk Assessment (TPRA) relies on large repositories of cybersecurity compliance questions used to assess external suppliers against standards such as ISO/IEC 27001 and NIST. In practice, not all questions are relevant for a specific supplier and selecting questions for a given assessment context remains a manual and time-consuming task. Existing question retrieval approaches based on lexical or semantic similarity can identify topically related questions, but they often fail to capture the underlying assessment intent, including control domain and evaluation scope. To address this limitation, we investigate whether an explicit semantic label space can improve intent-aware TPRA question selection. In particular, we separate label space discovery from large-scale label assignment. We start by discovering overlapping clusters of semantically similar questions and then exploit LLMs to assign unique labels for each cluster. Second, we propagate labels through k-nearest neighbors (kNN) for a larger-scale question annotation. Question retrieval is finally achieved by similarity measure of the query with respect to the extracted labels instead of the questions themselves. This reduces repeated LLM calls while preserving label consistency. Experimental results show that the proposed semi-supervised framework reduces labeling cost and runtime compared with per-question LLM annotation while maintaining label quality and improving efficiency. Furthermore, label-based retrieval achieves better alignment with cybersecurity control domains and assessment scope than similarity-based retrieval, highlighting the value of semantic labels as an intermediate representation.

arXiv:2603.23171v2 Announce Type: replace-cross Abstract: Large language models (LLMs) can be misused to reveal sensitive information, such as weapon-making instructions or writing malware. LLM providers rely on $\emph{monitoring}$ to detect and flag unsafe behavior during inference. An open security challenge is $\emph{adaptive}$ adversaries who craft attacks that simultaneously (i) evade detection while (ii) eliciting unsafe behavior. Adaptive attackers are a major concern as LLM providers cannot patch their security mechanisms, since they are unaware of how their models are being misused. We cast $\emph{robust}$ LLM monitoring as a security game, where adversaries who know about the monitor try to extract sensitive information, while a provider must accurately detect these adversarial queries at low false positive rates. Our work (i) shows that existing LLM monitors are vulnerable to adaptive attackers and (ii) designs improved defenses through $\emph{activation watermarking}$ by carefully introducing uncertainty for the attacker during inference. We find that $\emph{activation watermarking}$ outperforms guard baselines by up to $52\%$ under adaptive attackers who know the monitoring algorithm but not the secret key.