Papers

The decentralization of modern energy systems is transforming consumers into prosumers who continuously exchange data with aggregators, peers, and market operators. While such data is essential for peer-to-peer trading, demand response, and distributed forecasting, it can reveal sensitive household patterns and introduce privacy risks. Existing data sharing mechanisms rely on fixed policies or predefined differential privacy budgets, limiting their ability to adapt to variations in reliability, data sensitivity, and request purpose. As a result, prosumers rarely receive explanations for why a request is accepted, rejected, or modified, reducing trust and participation. To address these limitations, we propose X-NegoBox, an explainable negotiation framework for adaptive privacy budgeting and transparent decision making. Each prosumer data is managed locally within a private DataBox, where raw data remain confined. Incoming requests are processed by an Autonomous Privacy Budget Negotiation Protocol (APBNP), which determines an appropriate privacy budget based on trust, feature sensitivity, declared purpose, historical behavior, and risk-aware pricing. When needed, APBNP generates privacy-preserving counter-offers, such as reduced resolution or duration. An Explainable Agreement Layer (X-Contract) produces human- and machine-readable justifications for each decision. After agreement, requester code executes locally in a sandbox, and only sanitized outputs are shared. Experiments on realistic energy market settings show reduced privacy leakage, higher acceptance rates, and improved interpretability.

The Rowhammer vulnerability poses an increasing challenge with newer generations of DRAM and aggressive technology scaling. Existing mitigation techniques, such as Graphene, Twice, and Hydra, primarily rely on tracking activation counts for each row and issuing refreshes when a row reaches a predefined tracking threshold. However, these methods have inherent limitations, including inefficiencies in identifying rows genuinely at risk of bit flips. In this paper, we propose a novel framework called Rowhammer Vulnerability Count (RVC), which shifts the focus from activation count tracking to evaluating a row's actual vulnerability to bit flips. By selectively issuing refreshes only to rows on the verge of experiencing bit flips, RVC drastically reduces unnecessary refresh operations. We also demonstrate that prior works have incorrectly set tracking thresholds, leading to security flaws. Our evaluation shows that RVC achieves 95 - 99.99% improvement in mitigation induced refreshes when compared to Graphene, with no additional space overhead. Furthermore, RVC improves energy efficiency and reduces average LLC latency by up to 76.91%, making it a highly efficient and scalable solution for addressing Rowhammer in modern DRAM systems. These findings establish RVC as a superior approach for preventing Rowhammer, outperforming existing methods in both accuracy and efficiency.

Trusted Execution Environments (TEEs) on low-power microcontrollers (e.g., ARM TrustZone-M) enable isolation of Secure and Non-Secure software but still require both worlds to share resources, including interrupt controllers. In this model, real-time applications and real-time operating systems (RTOS-s) are executed in the Non-Secure sub-system, whereas the Secure sub-system is typically reserved for a small set of pre-defined security (e.g., cryptographic) operations referred to as trusted computing services. However, many RTOS-s rely on periodic interrupts (SysTicks) to advance their own notion of time (time-keeping), and the delivery of this interrupt is essential for preserving real-time behavior. On the other hand, the security of many trusted computing services requires atomicity vis-a-vis the Non-Secure sub-system (where the RTOS resides), precluding SysTick handling. This paper first characterizes this conflict and then introduces a Secure-driven time synchronization mechanism in which the Secure World measures elapsed time and compensates the Non-Secure RTOS by unobtrusively updating the RTOS time-keeping data structures with the appropriate number of missed ticks before re-enabling interrupts and resuming the execution of the Non-Secure system. This approach restores a consistent, monotonic notion of time across worlds and enables secure coexistence of trusted computing services and RTOS-s on microcontrollers. Importantly, the proposed approach requires no modifications to the underlying RTOS and yields no significant run-time overhead.

The convergence of Internet of Things (IoT) security and Zero Trust (ZT) principles is a trending topic, demanding a comprehensive, multi-perspective analysis. We present the first multivocal literature review (MLR) on this topic, combining 68 academic and 36 industrial studies. This comprehensive review identifies two complementary yet divergent perspectives: academia focuses on IoT compliance with ZT principles through IoT modifications, while industry prioritizes practical integration within existing ZT frameworks guided by NIST standards. The analysis reveals critical research gaps in socio-technical understanding, cost-benefit evaluation, and interdisciplinary collaboration, highlighting these as key directions for future research.

Auditing the semantic properties of proprietary data creates a fundamental tension: verification requires transparent access, while proprietary rights demand confidentiality. While Zero-Knowledge Proofs (ZKPs) ensure privacy, they are typically limited to precise algebraic constraints and are ill-suited for verifying qualitative, unstructured properties, such as the logic within a codebase. We propose Agentic Witnessing, a framework that moves verification from attested execution to attested reasoning. The system is composed of three agents: a Verifier (who wants to check properties of a dataset), a Prover (who owns the dataset) and an Auditor (that inspects the dataset). The Verifier is allowed to ask a limited number of simple binary true/false questions to the auditor. By isolating an LLM-based Auditor within a Trusted Execution Environment (TEE), the system enables the Verifier to query a Prover's private data via simple Boolean queries, without exposing the raw dataset. The Auditor uses the Model Context Protocol (MCP) to dynamically inspect the target dataset, producing a yes/no verdict accompanied by a cryptographic transcript: a signed hash chain binding the reasoning trace to both the original dataset and the TEE's hardware root of trust. We demonstrate this architecture by automating the artifact evaluation process for 21 peer-reviewed computer science papers with released codebases on GitHub (e.g. Does the codebase implement the system described in the paper?). We verified five high-level properties of these codebases described in the corresponding publications, treating the source code as private. Our results show that TEE-enabled agentic auditing provides a mechanism for privacy-preserving oversight, effectively decoupling qualitative verification from the need for data disclosure.

As LLM-driven agents advance in cybersecurity, Jeopardy CTF benchmarks are approaching saturation and cyber ranges, the natural next evaluation frontier, offer diminishing resistance under their current static design. We validate this observation by deploying an LLM-driven Advanced Persistent Threat (APT) agent across three tiers of increasingly realistic infrastructure (PRO Labs, MHBench, military-grade CYBER RANGES). To counteract this trend, we propose Dynamic Cyber Ranges: cyber range environments augmented with LLM-driven Defender agents that harden infrastructure, monitor for intrusions, and respond in real time. Across evaluated scenarios, Defender agents reduce attacker success to 0-55%, achieving complete prevention on multiple configurations. Since attacker and defender agents draw from the same underlying model capabilities, Dynamic Cyber Ranges preserve evaluation headroom as models improve. Notably, a smaller, specialized on-premise model (alias2-mini) matched the frontier model's defensive outcomes on multiple scenarios under identical, untuned prompts, and detected the attacker 10x faster on a complex enterprise scenario, suggesting that privacy-preserving on-premise models can serve as competent defenders against frontier-class attackers. The experiments further surface emergent agent behaviors, including scope expansion and prompt exfiltration, with implications for AI benchmark integrity and agentic system design.

Backdoored large language models (LLMs) exhibit attacker-specified behavior at inference time while retaining normal performance on benign inputs. Existing mitigations often require parameter updates and trusted clean data, or rely on auxiliary generation and repeated model execution, complicating deployment. Across diverse backdoor mechanisms, successful activations exhibit stronger tail concentration in attention over semantic-content tokens than benign inputs and unsuccessful trigger activations. This pattern provides a sample-internal control signal to selectively regulate suspicious attention dynamics. We propose TIARA, a tail-risk-informed attention rebalancing approach for inference-time LLM backdoor mitigation. TIARA filters structural attention sinks, aggregates sparse high-concentration events across rows and heads, and converts the risk signal into selective content-domain power smoothing and adaptive attention-mass reallocation. A constrained reconstruction writes valid attention distributions back before value aggregation. TIARA requires no parameter updates, auxiliary generation, additional target-model passes, or deployment-time clean reference sets. We evaluate TIARA across four backdoor paradigms on dense, reasoning-oriented, and sparse mixture-of-experts LLMs. Across three model families, TIARA reduces average macro ASR to 11.5%, outperforming the strongest no-update inference-time baseline by 7.2 percentage points while limiting clean-task degradation to at most 3.8 percentage points. Under standardized profiling, TIARA adds 12.9% end-to-end latency over a matched eager-attention baseline; the current unfused path is 23.3% slower than fused SDPA. Overall, TIARA establishes sample-conditional attention rebalancing as a practical inference-time control layer for mitigating attention-concentrated LLM backdoors.

Spotting encryption loops in binary-only ransomware is a critical reverse engineering task. Since the existence of avalanche effect, an intrinsic characteristic of any secure encryption algorithms, is unavoidable during a victim data encryption attack, it is a very promising direction to spot encryption loops through avalanche effect detection. Unfortunately, no existing work in this direction ensures that the being-checked effect is the avalanche effect itself. Although CipherXRay is inspired by avalanche effect, it only checks whether a "ripple effect" (i.e., a necessary but non-sufficient condition) of avalanche effect exists, allowing a straightforward counterattack to succeed. In this work, we present a new approach that checks the avalanche effect itself. Because the detection is conducted in adversarial settings (e.g., the ransomware author may obfuscate the code), a viable approach must tolerate inaccurate input & output identification and must be resilient to adversarial evasion. These challenges are addressed by a novel record-and-replay detection mechanism that takes advantage of the statistical guarantees provided by the Shapiro-Wilk normality test. The experimental results show that our approach achieves 0.0% false negative rate and 1.1% false positive rate. When our tool is employed to reverse engineer real-world ransomware samples, it succeeds in analyzing all the ransomware samples selected from ten representative families.

Large Language Model (LLM) agents are increasingly used to automate complex workflows, but integrating untrusted external data with privileged execution exposes them to severe security risks, particularly direct and indirect prompt injection. Existing defenses face significant challenges in balancing security with utility, often encountering a trade-off where rigorous protection leads to over-defense, or where subtle indirect injections bypass detection. Drawing inspiration from operating system virtualization, we propose AgentVisor, a novel defense framework that enforces semantic privilege separation. AgentVisor treats the target agent as an untrusted guest and intercepts tool calls via a trusted semantic visor. Central to our approach is a rigorous audit protocol grounded in classic OS security primitives, designed to systematically mitigate both direct and indirect injection attacks. Furthermore, we introduce a one-shot self-correction mechanism that transforms security violations into constructive feedback, enabling agents to recover from attacks. Extensive experiments show that AgentVisor reduces the attack success rate to 0.65%, achieving this strong defense while incurring only a 1.45% average decrease in utility relative to the No Defense scenario, demonstrating superior performance compared to existing defense methods.

Cryptographic API misuse represents a critical vulnerability class that undermines the security foundations of modern software. Yet, it remains largely unexplored in Go despite its dominance in security-critical infrastructure. This paper presents the first comprehensive study of cryptographic API misuse detection in Go, identifying and analyzing 4 state-of-the-art tools (CodeQL, Gopher, Gosec, and Snyk Code) and establishing a consolidated taxonomy of 14 relevant misuse classes. Through an experimental evaluation of 328 security-critical open-source Go projects, we discovered 7,473 cryptographic API misuses, providing insights into the prevalence and distribution of these vulnerabilities. Our systematic comparison reveals significant variations in misuse coverage, with immediate practical implications for security engineers and long-term implications for research in this domain.

Runtime monitoring of stochastic systems must distinguish nominal distributional relaxation from regime departure while controlling repeated-test false alarms under explicit validity assumptions. This paper links relative-entropy dissipation, information geometry, and sequential inference in a bounded first-passage monitoring architecture. For reversible Fokker–Planck dynamics, relative entropy to an invariant density is non-increasing; under exogenous forcing, its derivative decomposes into nominal dissipation and an information-space forcing term. The runtime layer uses Gaussian window surrogates, nominal-relative covariance shrinkage, a coordinate-consistent relative precision diagnostic, and randomized conformal ranks aggregated by a mixture power-martingale process. Analytical Ornstein–Uhlenbeck validation gives zero positive nominal Kullback–Leibler increments, forcing-identity residuals below 3.31 x 10^-6, and coordinate-invariance errors at numerical roundoff. On NSL-KDD, the monitor yields 0/100 alarms on internal nominal streams but 63/100 on official test-normal streams; post-change detection is 99.0% for seen and 98.53% for test-only attack types with median one-window delay. On UNSW-NB15, internal-null alarms are 0/100, whereas official test-normal alarms rise to 90/100; post-change detection is 81.33%, with 18.67% pre-change alarms. In these evaluations, calibration transport emerges as a major deployment constraint. No universal benchmark superiority, causal inference, or physical-work interpretation is claimed.

Large (vision-)language models exhibit remarkable capability but remain highly susceptible to jailbreaking. Existing safety training approaches aim to have the model learn a refusal boundary between safe and unsafe, based on the user's intent. It has been found that this binary training regime often leads to brittleness, since the user intent cannot reliably be evaluated, especially if the attacker obfuscates their intent, and also makes the system seem unhelpful. In response, frontier models, such as GPT-5, have shifted from refusal-based safeguards to safe completion, that aims to maximize helpfulness while obeying safety constraints. However, safe completion could be exploited when a user pretends their intention is benign. Specifically, this intent inversion would be effective in multi-turn conversation, where the attacker has multiple opportunities to reinforce their deceptively benign intent. In this work, we introduce a novel multi-turn jailbreaking method that exploits this vulnerability. Our approach gradually builds conversational trust by simulating benign-seeming intentions and by exploiting the consistency property of the model, ultimately guiding the target model toward harmful, detailed outputs. Most crucially, our approach also uncovered an additional class of model vulnerability that we call para-jailbreaking that has been unnoticed up to now. Para-jailbreaking describes the situation where the model may not reveal harmful direct reply to the attack query, however the information that it reveals is nevertheless harmful. Our contributions are threefold. First, it achieves high success rates against frontier models including GPT-5-thinking and Claude-Sonnet-4.5. Second, our approach revealed and addressed para-jailbreaking harmful output. Third, experiments on multimodal VLM models showed that our approach outperformed state-of-the-art models.

As large language models (LLMs) are increasingly deployed in high-stakes and operational settings, evaluation strategies based solely on aggregate accuracy are often insucient to characterize system reliability. This study proposes a thermodynamic inspired modeling framework for analyzing the stability of LLM outputs under conditions of uncertainty and perturbation. The framework introduces a composite stability score that integrates task utility, entropy as a measure of external uncertainty, and two internal structural proxies: internal integration and aligned reective capacity. Rather than interpreting these quantities as physical variables, the formulation is intended as an interpretable abstraction that captures how internal structure may modulate the impact of disorder on model behavior. Using the IST-20 benchmarking protocol and associated metadata, we analyze 80 modelscenario observations across four contemporary LLMs. The proposed formulation consistently yields higher stability scores than a reduced utilityentropy baseline, with a mean improvement of 0.0299 (95% CI: 0.02470.0351). The observed gain is more pronounced under higher entropy conditions, suggesting that the framework captures a form of nonlinear attenuation of uncertainty. We do not claim a fundamental physical law or a complete theory of machine ethics. Instead, the contribution of this work is a compact and interpretable modeling perspective that connects uncertainty, performance, and internal structure within a unied evaluation lens. The framework is intended to complement existing benchmarking approaches and to support ongoing discussions in AI safety, reliability, and governance.

Mobile apps frequently request excessive data access, raising significant privacy concerns. While regulations like GDPR emphasize data minimization, they provide limited guidance on concretely defining and enforcing necessary data access. Existing regulatory mechanisms primarily rely on expert-driven audits that face challenges in scalability, neutrality, and alignment with user expectations. In this paper, we propose a novel paradigm–democratizing privacy assessment, inspired by prior work on user-centric privacy perceptions–which repositions users as active evaluators in the privacy auditing process, recognizing that user perceptions of data usage play a crucial role in assessing the appropriateness and necessity of data access. To operationalize this paradigm, we introduce DePRa, a prototype system developed through participatory design, featuring contextual explanation provision, category-based representative selection, an intuitive rating interface, and preference-based rating adjustment. We evaluated DePRa with 200 everyday mobile app users, analyzing how effectively it captures user opinions on sensitive data access, comparing their privacy ratings with expert assessments, and exploring risk preference-based score calibration. Our findings show the feasibility and promise of democratized privacy assessment, highlighting its potential to complement expert auditing and support inclusive privacy evaluation.

Industrial Control Systems (ICS) integrate computing, physical processes, and communication to operate critical infrastructures such as power grids, water treatment plants, and oil and gas facilities. As ICS become increasingly targeted by cyberattacks, timely and reliable anomaly diagnosis is essential for protecting operational safety. However, existing ICS anomaly detection approaches face practical limitations: supervised methods require extensive labeled attack data and suffer from class imbalance, while model-based detectors often lack the ability to provide deep insight into the root causes of anomalies, leading to elevated false alarms and making it difficult for operators to initiate a timely response. In this work, we propose a system-aware unsupervised framework for ICS anomaly diagnosis that combines lightweight online detection with contextual explanation. The system identifies deviations from observed normal behaviors without prior knowledge of system topology. To support actionable response, we further concatenate a contextual digital twin augmented with an Large Language Model (LLM) to enhance interpretability, which translates detection evidence into grounded diagnostic hypotheses and verification steps for operators. Experiments on public ICS benchmarks demonstrate that the proposed framework achieves real-time detection efficiency and provides consistent, interpretable anomaly diagnoses, enabling low-latency warning and practical deployment in complex industrial environments.

Autonomous AI agents deployed on platforms such as OpenClaw face prompt injection, memory poisoning, supply-chain attacks, and social engineering, yet existing defences address only the platform perimeter, leaving the agent's own threat judgement entirely untrained. We present ClawdGo, a framework for endogenous security awareness training: we teach the agent to recognise and reason about threats from the inside, at inference time, with no model modification. Four contributions are introduced: TLDT (Three-Layer Domain Taxonomy) organises 12 trainable dimensions across Self-Defence, Owner-Protection, and Enterprise-Security layers; ASAT (Autonomous Security Awareness Training) is a self-play loop where the agent alternates attacker, defender, and evaluator roles under weakest-first curriculum scheduling; CSMA (Cross-Session Memory Accumulation) compounds skill gains via a four-layer persistent memory architecture and Axiom Crystallisation Promotion (ACP); and SACP (Security Awareness Calibration Problem) formalises the precision-recall tradeoff introduced by endogenous training. Live experiments show weakest-first ASAT raises average TLDT score from 80.9 to 96.9 over 16 sessions, outperforming uniform-random scheduling by 6.5 points and covering 11 of 12 dimensions. CSMA retains the full gain across sessions; cold-start ablation recovers only 2.4 points, leaving a 13.6-point gap. E-mode generates 32 TLDT-conformant scenarios covering all 12 dimensions. SACP is observed when a heavily trained agent classifies a legitimate capability assessment as prompt injection (30/160).

Agent systems increasingly execute rather than advise. When an AI agent queries regulated data, invokes effectful tools, and mutates persistent state, correctness is not captured by whether a terminal output looks plausible. The operative questions are whether each step was authorized under a contract, whether the recorded history is tamper-evident, and whether the trajectory can be reconstructed deterministically. We formalize this as runtime proof of execution. An execution is a triple $x = (C, T, R)$: a contract $C$, an Execution Causal Event Stream (ECES) $T$, and a replay context $R$. A well-formedness predicate and five validator-checkable invariants form the PoE validity predicate. Five semantic guarantees describe authorization, path compliance, null effect on deny, history integrity, and replayability. We prove soundness under explicit cryptographic and deployment assumptions: any PPT adversary that produces a PoE-valid execution violating a semantic guarantee yields a signature forgery, a hash collision, or a quantified deployment-failure event. The Prime Execution Model (PEM) separates planning, enforcement, effect, and recordkeeping into distinct authority planes; a lemma reduces trace completeness to Effector-exclusive credentialing. An Execution Attestation Certificate is issued only when PoE = 1. In a single-node TypeScript prototype, PoE adds approximately 2.7 ms on a minimal flow and 4.4% overhead on concurrent batch workloads; a standard eight-event trace compresses to approximately 1.1 KB; injected Gateway-bypass and trace-mutation attacks are rejected. PoE does not replace consensus, TEEs, or zkVMs; it binds authorization, effect, history, and replay into a single runtime-checkable object so that governed execution becomes attestable under contract.

Threat modeling for cyber-physical systems (CPS) remains a largely manual exercise. This project presents SMSI (System Model Security Inference), a hybrid neuro-symbolic pipeline that starts from a SysML architecture model and produces a prioritized list of NIST 800-53 security controls. The prototype has three main stages: a deterministic parser mapping system components to vulnerabilities via the NVD; a family of retrieval and classification models linking vulnerabilities to MITRE ATT&CK techniques; and a control recommender. We explore three approaches for CVE-to-ATT&CK mapping: a supervised classifier using fine-tuned SecureBERT+, retrieval-based dense encoders, and a zero-shot LLM approach using Gemma-4 26B. We validate the pipeline on a healthcare IoT gateway with nine software components. For the ATT&CK-to-NIST stage, pretrained SecureBERT achieves the highest control retrieval scores, demonstrating that dense embeddings provide a strong basis for automated control recommendation.

LLM-powered applications routinely embed secrets in system prompts, yet models can be tricked into revealing them. We built an adaptive attacker that evolves its strategies over hundreds of rounds and tested it against nine defense configurations across more than 20,000 attacks. Every defense that relied on the model to protect itself eventually broke. The only defense that held was output filtering, which checks the model's responses via hardcoded rules in separate application code before they reach the user, achieving zero leaks across 15,000 attacks. These results demonstrate that security boundaries must be enforced in application code, not by the model being attacked. Until such defenses are verified by tools like Swept AI, AI systems handling sensitive operations should be restricted to internal, trusted personnel.

Rootkits are among the most elusive types of malware, capable of bypassing traditional static analysis methods due to their metamorphic behavior. Signature-based detection techniques struggle against these threats, necessitating a shift toward dynamic analysis approaches. We propose SeqShield, a behavior-based rootkit detection approach designed specifically for the Windows OS, leveraging API call sequences for dynamic behavior analysis. Instead of relying on static signatures, SeqShield examines the execution patterns of API calls, which inherently reflect malicious intent. Analyzing API sequences, we can effectively identify rootkit-like behavior. We also employed a metamorphic code engine to generate 10X mutated variants of rootkits, demonstrating their obfuscation strategies. SeqShield applies n-gram analysis to extract bigram and trigram features from these API call sequences, enabling effective detection of rootkit-like activity. Among the models tested, Random Forest achieves the highest accuracy of 97.27% (bigram) and 96.17% (trigram). To optimize performance and decrease the dimension, we apply feature importance ranking using the Gini Impurity Index, iteratively selecting the most significant features. The optimized lower-dimensional feature matrix significantly enhances detection efficiency without sacrificing accuracy. Using the optimized feature set, our approach achieves 96.72% accuracy for bigrams and 97.81% accuracy for trigrams.

This paper extends the Classification Error Gauge (x-CEG) framework, originally developed for measuring the privacy-utility trade-off in tabular datasets, to privacy auditing of Large Language Models (LLMs). We propose LLM-CEG, a systematic framework that employs membership inference attack (MIA) success rates as an empirical privacy gauge and model perplexity as a utility gauge, iteratively adjusting differential privacy parameters until both thresholds are jointly satisfied. A proof-of-concept prototype fine-tunes DistilGPT-2 on a synthetic clinical PII dataset under four privacy regimes using DP-SGD. Results indicate that DP-SGD reduces MIA attacker advantage by 71.5% while simultaneously improving out-of-distribution utility by 47-50% relative to the overfitted baseline, suggesting that differential privacy may act as implicit regularization under narrow fine-tuning conditions. We further extend the SIED engineering framework to the LLM context as LLM-SIED, providing an auditable, regulator-aligned process for privacy-compliant LLM deployment.

With the wide adoption of personal AI assistants such as OpenClaw, privacy leakage in user interaction contexts with large language model (LLM) agents has become a critical issue. Existing privacy attacks against LLMs primarily target training data, while research on inference-time contextual privacy risks in LLM agent memory remains limited. Moreover, prior methods often incur high attack costs, requiring multiple queries or relying on white-box assumptions, which limits their practicality in real-world deployments. To address these issues, we propose a training-free privacy extraction attack targeting LLM agent memory, which we name Spore. Spore is compatible with both black-box and gray-box settings. In the black-box setting, Spore can efficiently extract a small candidate set via a single query to recover the original private information. In the gray-box setting, Spore allows the attacker to leverage multi-ranked tokens for more accurate and faster privacy extraction. We provide an information-theoretic analysis of Spore and show that it achieves high query efficiency with substantial per query information leakage. Experiments on multiple frontier LLMs show that Spore outperforms attack success rate over existing state-of-the-art (SOTA) schemes. It also maintains low attack cost and remains stable across different model parameter settings. We further evaluate the robustness of Spore against existing defense mechanisms. Our results show that Spore consistently bypasses both detection and strong safety alignment, demonstrating resilient performance in diverse defensive settings and real-world safety threats.

Line current differential relays (LCDRs) are measurement-driven relays that rely on time-synchronized multi-phase current waveforms to infer internal faults in AC and DC power networks. In inverter-based microgrids, however, the increasing reliance on digitally communicated measurements exposes LCDRs to false-data injection attacks (FDIAs), in which adversaries manipulate remote measurement streams to create protection-triggering yet physically inconsistent current trajectories. This paper addresses this emerging measurement integrity problem by introducing a measurement integrity validation scheme that operates as a supervisory instrumentation layer for modern LCDRs. The proposed scheme interprets short windows of synchronized instantaneous current measurements recorded during relay operation and assesses their physical consistency to distinguish genuine fault-induced trajectories from cyber-manipulated measurement streams. A recurrent neural network is trained offline using only relay-available current measurements and exploits the temporal structure of differential current waveforms, which remains informative in inverter-dominated systems where current magnitude is no longer a reliable observable. The method requires no additional sensors, auxiliary protection elements, or prior knowledge of network topology, and is applicable to both AC and DC LCDRs without structural modification. The proposed measurement validation scheme is evaluated on an islanded inverter-based microgrid under a comprehensive set of fault and FDIA scenarios, demonstrating high detection accuracy while preserving relay dependability. Hardware-in-the-loop validation using an OPAL-RT real-time simulator confirms that the scheme satisfies protection timing constraints and can operate in real time under realistic operating conditions.

Rényi Pufferfish Privacy (RPP) provides a Rényi divergence-based privacy framework for correlated data, but existing $\infty$-Wasserstein mechanisms are often conservative and sacrifice data utility. We study Gaussian mechanisms for RPP under Gaussian and Gaussian-mixture priors. For single Gaussian priors, we derive the exact Rényi divergence after Gaussian perturbation, obtain a relaxed closed-form sufficient condition for $(\alpha,\epsilon)$-RPP, and characterize the monotonicity of the calibrated noise with respect to the privacy budget $\epsilon$ and the Rényi order $\alpha$. To handle more general non-Gaussian and multimodal priors, we approximate secret-conditioned outputs with Gaussian mixture models and introduce an optimal-transport-based sufficient condition for RPP. Experiments on three UCI datasets with statistical (RAW, MEAN) and model-output (BNN, GP) queries show that our prior-aware mechanisms consistently require less noise than a recent RPP additive-noise baseline, achieving an average noise reduction of 48.9%. These results show that our mechanisms can substantially improve the privacy-utility trade-off under RPP.

Recent evidence suggests that frontier AI systems can exhibit agentic misalignment, generating and executing harmful actions derived from internally constructed goals, even without explicit user requests. Existing mitigation methods, such as Reinforcement Learning from Human Feedback (RLHF) and constitutional prompting, operate primarily at the model level and provide only probabilistic safety guarantees. We propose the Policy-Execution-Authorization (PEA) architecture, a "separation-of-powers" design that enforces safety at the system level. PEA decouples intent generation, authorization, and execution into independent, isolated layers connected via cryptographically constrained capability tokens. We present five core contributions: (C1) an Intent Verification Layer (IVL) for ensuring capability-intent consistency; (C2) Intent Lineage Tracking (ILT), which binds all executable intents to the originating user request via cryptographic anchors; (C3) Goal Drift Detection, which rejects semantically divergent intents below a configurable threshold; (C4) an Output Semantic Gate (OSG) that detects implicit coercion using a structured $K \times I \times P$ threat calculus (Knowledge, Influence, Policy); and (C5) a formal verification framework proving that goal integrity is maintained even under adversarial model compromise. By shifting agent alignment from a behavioral property to a structurally enforced system constraint, PEA provides a robust foundation for the governance of autonomous agents.

Vehicle diagnostics has become essential for detecting in-vehicle errors and ensuring safety. While the Unified Diagnostic Services (UDS) protocol is widely adopted for diagnostic operations, it relies on the ISO 15765-2 standard as the transport protocol over the Controller Area Network (CAN), which was designed without inherent security considerations. In this paper, we identify eight novel attack scenarios that exploit specific transport layer mechanisms in the ISO 15765-2 standard, including Flow Control manipulation, Sequence Number violations, and error handling abuses. We evaluate these attacks on a real passenger vehicle using two distinct diagnostic tools to demonstrate their practical impact. Our results confirm that three of these attack scenarios successfully induce denial of diagnostic services, leading to abnormal diagnostic results such as concealed faults and manipulated sensor readings. These findings highlight critical vulnerabilities that can deceive technicians and drivers, potentially exposing vehicles to significant safety risks.

The widespread open-sourcing of advanced recommendation algorithms and the rising threat of model extraction attacks have made safeguarding the intellectual property of recommender systems an imperative task. While watermarking serves as a potent defense, existing methods primarily rely on forcing models to memorize pre-defined interaction patterns. Such memorization-based approaches often require excessive synthetic data injection and are vulnerable to removal attacks due to their detectable statistical deviations from natural user behavior. To address these limitations, we propose GREW, a novel Green-REd Watermarking framework for recommender systems. GREW leverages a secret key to partition the item space into "green" items for soft promotion and "red" items as anchors, thereby shifting the paradigm from fragile memorization to a stealthy, key-controlled output bias. By integrating watermark signals directly into the intrinsic ranking process, GREW employs three recommendation-tailored modules: (1) Semantic-Consistent Hashing, which utilizes the secret key to cluster green items for performance-aware stealthiness; (2) Decision-Aligned Masking, which confines signal injection to the competitive item subset to preserve ranking logic; and (3) Confidence-Aware Scaling, which dynamically modulates injection intensity based on model uncertainty. Ownership verification is performed via statistical hypothesis testing on aggregated black-box outputs, enabled by the keyed re-partitioning of the item space. Experiments on multiple base models demonstrate that GREW achieves strong ownership verification and robustness against extraction attacks compared to existing baselines while requiring no data injection. Our code is available at https://github.com/Loche2/GREW.

Organizations usually calibrate phishing detectors on public corpora, because privacy rules make their own labeled legitimate mail hard to collect. This practice breaks the false-positive budget of trained detectors: calibrated for a 1% false-positive rate (FPR) on older public mail and deployed on DataPhish 2025, they realize 20.2-22.9% FPR, 9-94 times the budget across three calibration sources. The cause is calibration transfer, not ranking: the classifiers separate the deployment mail well, but their thresholds do not transfer. We introduce CyberCane, a privacy-bounded neuro-symbolic detector that scores email only by similarity to known phishing in a redacted, training-only index, plus deterministic rules, and calibrates verdicts to an explicit FPR budget. Because legitimate mail never enters its score, only its threshold depends on the calibration source. Under the same shift CyberCane realizes 5.4% FPR at 86.4% recall, reducing the overrun 2.7-6.1 times across sources, seeds, a second shift family, and a local embedder. A segment-level variant, scoring an email's most phishing-like part, lowers the realized FPR a further 1.2-2.7 times at equal ranking quality and withstands benign padding. CyberCane screens an email locally in 0.20 s, sending no text out; a 70B zero-shot LLM ranks better at strict FPR but is 14-24 times slower and must receive each email. We also show that LLM rewriting does not evade embedding-based detectors and that a random split of DataPhish 2025 leaks 96% of its test set. All protocols were fixed in advance, and the released splits and code let a deployer measure realized FPR before trusting a budget.

Conflict-free replicated data types (CRDTs) and the local-first concept are increasingly employed not only in small-scale collaboration systems among few users who trust each other, but also in large-scale systems, like Matrix for instant messaging and Keyhive for collaborative documents. Since mutual trust is no longer warranted, these systems require Byzantine fault tolerance and fine-grained access control. As of today, Matrix and Keyhive pair an informal specification with an unverified reference implementation. In this work, we follow a bottom-up approach towards constructing formally verified authorization algorithms for Byzantine fault-tolerant local-first systems. As intermediate target for formal verification, we contribute semantics and invariants of a replicated data type for managing simplified collaboration groups, based on capabilities for access control and hash chronicles for replication. To enable future integration into local-first systems like Matrix and Keyhive, we strive for accessibility to system engineers by using the Rust programming language for formal specification, verification, and implementation, enabled by the Verus framework using the Z3 theorem prover at zero runtime cost. We report on our experience and preliminary results following this approach, and discuss next steps towards scaling up access control expressiveness. Whether this approach can be scaled up to the complexity of real-world local-first access control systems like Matrix or Keyhive remains future work, but our findings demonstrate the potential of system-oriented formal verification with Verus.

The aviation industry faces significant vulnerabilities from both physical and cybersecurity threats, highlighting the urgent need for enhanced cybersecurity measures amid increasingly sophisticated attacks. This paper systematically reviews emerging threats at airports, analyzing real-world incidents and relevant literature while mapping risks to the MITRE ATT&CK Matrix, a widely recognized knowledge base for categorizing cyberattack tactics, techniques, and procedures. This is the first to apply the MITRE Matrix to airport security risks, offering a novel approach to understanding and mitigating these challenges. Building on this analysis, the paper advocates for modern cybersecurity defense models, emphasizing Cybersecurity Frameworks and Zero Trust Architecture, as well as critical measures for supply chain risk management and strategies to mitigate ransomware and DoS attacks. Our analysis provides insights into vulnerabilities and actionable recommendations, serving as a comprehensive guide for aviation stakeholders to strengthen defenses against evolving cybersecurity threats.