Papers

Malware often uses obfuscation to hinder security analysis. Among these techniques, virtualization-based obfuscation is particularly strong because it protects programs by translating original instructions into attacker-defined virtual machine (VM) bytecode, producing long and complex code that is difficult to analyze and deobfuscate. This paper aims to identify the structural components of virtualization-based obfuscation through static analysis. By examining the execution model of obfuscated code, we define and detect the key elements required for deobfuscation-namely the dispatch routine, handler blocks, and the VM region-using LLVM IR. Experimental results show that, in the absence of compiler optimizations, the proposed LLVM Pass successfully detects all core structures across major virtualization options, including switch, direct, and indirect modes.

While many resource-constrained networks, such as Internet of Things (IoT) and Internet of Vehicles (IoV), are inherently distributed, the majority still rely on central servers for fast authentication and data sharing. Blockchain-based solutions offer decentralized alternatives but often struggle to meet the stringent latency requirements of real-time applications. Even with the rollout of 5G, network latency between servers and peers remains a significant challenge. To address this, we introduce SWORD, a novel offline-first authentication and data-sharing scheme designed specifically for resource-constrained networks. SWORD utilizes a proximity-based clustering approach to enable offline authentication and data sharing, ensuring low-latency, secure operations even in intermittently connected scenarios. Our experimental results show that SWORD outperforms traditional blockchain-based solutions while offering similar resource efficiency and authentication latency to central-server-based solutions. Additionally, we provide a comprehensive security analysis, demonstrating that SWORD is resilient against spoofing, impersonation, replay, and man-in-the-middle attacks.

The increasing prevalence of malicious Portable Document Format (PDF) files necessitates robust and comprehensive feature extraction techniques for effective detection and analysis. This work presents a unified framework that integrates graph-based, structural, and metadata-driven analysis to generate a rich feature representation for each PDF document. The system extracts text from PDF pages and constructs undirected graphs based on pairwise word relationships, enabling the computation of graph-theoretic features such as node count, edge density, and clustering coefficient. Simultaneously, the framework parses embedded metadata to quantify character distributions, entropy patterns, and inconsistencies across fields such as author, title, and producer. Temporal features are derived from creation and modification timestamps to capture behavioral signatures, while structural elements including, object streams, fonts, and embedded images, are quantified to reflect document complexity. Boolean flags for potentially malicious PDF constructs (e.g., JavaScript, launch actions) are also extracted. Together, these features form a high-dimensional vector representation (170 dimensions) that is well-suited for downstream tasks such as malware classification, anomaly detection, and forensic analysis. The proposed approach is scalable, extensible, and designed to support real-world PDF threat intelligence workflows.6

Modern Voice Control Systems (VCS) rely on the collaboration of Automatic Speech Recognition (ASR) and Speaker Recognition (SR) for secure interaction. However, prior adversarial attacks typically target these tasks in isolation, overlooking the coupled decision pipeline in real-world scenarios. Consequently, single-task attacks often fail to pose a practical threat. To fill this gap, we first utilize gradient analysis to reveal that ASR and SR exhibit no inherent conflicts. Building on this, we propose Dual-task Universal Adversarial Perturbation (DUAP). Specifically, DUAP employs a targeted surrogate objective to effectively disrupt ASR transcription and introduces a Dynamic Normalized Ensemble (DNE) strategy to enhance transferability across diverse SR models. Furthermore, we incorporate psychoacoustic masking to ensure perturbation imperceptibility. Extensive evaluations across five ASR and six SR models demonstrate that DUAP achieves high simultaneous attack success rates and superior imperceptibility, significantly outperforming existing single-task baselines.

Cellular networks are critical infrastructure supporting billions of worldwide users and safety- and mission-critical services. Vulnerabilities in cellular networks can therefore cause service disruption, privacy breaches, and broad societal harm, motivating growing efforts to analyze 3GPP specifications that define required device and operator behavior. While large language models (LLMs) have demonstrated the capability for reading technical documents, cellular specifications impose unique challenges: faithful interpretation of normative language, reasoning across cross-referenced clauses, and verifiable conclusions grounded in multimodal evidence such as tables and figures. To address these challenges, we propose CellSpecSec-ARI, a unified Adapt-Retrieve-Integrate framework for systematic understanding and standard-driven security analysis of 3GPP specifications; CellularSpecSec-Bench, a staged benchmark, containing newly constructed high-quality datasets with expert-verified and corrected subsets from prior open-source resources. Together, they establish an accessible and reproducible foundation for quantifying progress in specification understanding and security reasoning in the cellular network security domain.

Federated real-time object detection using transformers in Intelligent Transportation Systems (ITS) faces three major challenges: (1) missing-class non-IID data heterogeneity from geographically diverse traffic environments, (2) latency constraints on edge hardware for high-capacity transformer models, and (3) privacy and security risks from untrusted client updates and centralized aggregation. We propose BlockSecRT-DETR, a BLOCKchain-SECured Real-Time Object DEtection TRansformer framework for ITS that provides a decentralized, token-efficient, and privacy-preserving federated training solution using RT-DETR transformer, incorporating a blockchain-secured update validation mechanism for trustworthy aggregation. In this framework, challenges (1) and (2) are jointly addressed through a unified client-side design that integrates RT-DETR training with a Token Engineering Module (TEM). TEM prunes low-utility tokens, reducing encoder complexity and latency on edge hardware, while aggregated updates mitigate non-IID data heterogeneity across clients. To address challenge (3), BlockSecRT-DETR incorporates a decentralized blockchain-secured update validation mechanism that enables tamper-proof, privacy-preserving, and trust-free authenticated model aggregation without relying on a central server. We evaluated the proposed framework under a missing-class Non-IID partition of the KITTI dataset and conducted a blockchain case study to quantify security overhead. TEM improves inference latency by 17.2% and reduces encoder FLOPs by 47.8%, while maintaining global detection accuracy (89.20% mAP@0.5). The blockchain integration adds 400 ms per round, and the ledger size remains under 12 KB due to metadata-only on-chain storage.

Digital lending applications, commonly referred to as loan apps, have become a primary channel for microcredit in emerging markets. However, many of these apps demand excessive permissions and misuse sensitive user data for coercive debt-recovery practices, including harassment, blackmail, and public shaming that affect both borrowers and their contacts. This paper presents the first cross-country measurement of loan app compliance against both national regulations and Google's Financial Services Policy. We analyze 434 apps drawn from official registries and app markets from Indonesia, Kenya, Nigeria, Pakistan, and the Philippines. To operationalize policy requirements at scale, we translate policy text into testable permission checks using LLM-assisted policy-to-permission mapping and combine this with static and dynamic analyses of loan apps' code and runtime behavior. Our findings reveal pervasive non-compliance among approved apps: 141 violate national regulatory policy and 147 violate Google policy. Dynamic analysis further shows that several apps transmit sensitive data (contacts, SMS, location, media) before user signup or registration, undermining informed consent and enabling downstream harassment of borrowers and third parties. Following our disclosures, Google removed 93 flagged apps from Google Play, representing over 300M cumulative installs. We advocate for adopting our methodology as a proactive compliance-monitoring tool and offer targeted recommendations for regulators, platforms, and developers to strengthen privacy protections. Overall, our results highlight the need for coordinated enforcement and robust technical safeguards to ensure that digital lending supports financial inclusion without compromising user privacy or safety.

Algorithmic stablecoins promise decentralized monetary stability by maintaining a target peg through programmatic reserve management. Yet, their reserve controllers remain vulnerable to regime-blind optimization, calibrating risk parameters on fair-weather data while ignoring tail events that precipitate cascading failures. The March 2020 Black Thursday collapse, wherein MakerDAO's collateral auctions yielded $8.3M in losses and a 15% peg deviation, exposed a critical gap: existing models like SAS systematically omit extreme volatility regimes from covariance estimates, producing allocations optimal in expectation but catastrophic under adversarial stress. We present MVF-Composer, a trust-weighted Mean-Variance Frontier reserve controller incorporating a novel Stress Harness for risk-state estimation. Our key insight is deploying multi-agent simulations as adversarial stress-testers: heterogeneous agents (traders, liquidity providers, attackers) execute protocol actions under crisis scenarios, exposing reserve vulnerabilities before they manifest on-chain. We formalize a trust-scoring mechanism T: A -> [0,1] that down-weights signals from agents exhibiting manipulative behavior, ensuring the risk-state estimator remains robust to signal injection and Sybil attacks. Across 1,200 randomized scenarios with injected Black-Swan shocks (10% collateral drawdown, 50% sentiment collapse, coordinated redemption attacks), MVF-Composer reduces peak peg deviation by 57% and mean recovery time by 3.1x relative to SAS baselines. Ablation studies confirm the trust layer accounts for 23% of stability gains under adversarial conditions, achieving 72% adversarial agent detection. Our system runs on commodity hardware, requires no on-chain oracles beyond standard price feeds, and provides a reproducible framework for stress-testing DeFi reserve policies.

The increase in the number of Internet of Things (IoT) devices has tremendously increased the attack surface of cyber threats thus making a strong intrusion detection system (IDS) with a clear explanation of the process essential towards resource-constrained environments. Nevertheless, current IoT IDS systems are usually traded off with detection quality, model elucidability, and computational effectiveness, thus the deployment on IoT devices. The present paper counteracts these difficulties by suggesting an explainable AI (XAI) framework based on an optimized Decision Tree classifier with both local and global importance methods: SHAP values that estimate feature attribution using local explanations, and Morris sensitivity analysis that identifies the feature importance in a global view. The proposed system attains the state of art on the test performance with 99.91% accuracy, F1-score of 99.51% and Cohen Kappa of 0.9960 and high stability is confirmed by a cross validation mean accuracy of 98.93%. Efficiency is also enhanced in terms of computations to provide faster inferences compared to those that are generalized in ensemble models. SrcMac has shown as the most significant predictor in feature analyses according to SHAP and Morris methods. Compared to the previous work, our solution eliminates its major drawback lack because it allows us to apply it to edge devices and, therefore, achieve real-time processing, adhere to the new regulation of transparency in AI, and achieve high detection rates on attacks of dissimilar classes. This combination performance of high accuracy, explainability, and low computation make the framework useful and reliable as a resource-constrained IoT security problem in real environments.

Image transmission and processing systems in resource-critical applications face significant challenges from adversarial perturbations that compromise mission-specific object classification. Current robustness testing methods require excessive computational resources through exhaustive frame-by-frame processing and full-image perturbations, proving impractical for large-scale deployments where massive image streams demand immediate processing. This paper presents DDSA (Dual-Domain Strategic Attack), a resource-efficient adversarial robustness testing framework that optimizes testing through temporal selectivity and spatial precision. We introduce a scenario-aware trigger function that identifies critical frames requiring robustness evaluation based on class priority and model uncertainty, and employ explainable AI techniques to locate influential pixel regions for targeted perturbation. Our dual-domain approach achieves substantial temporal-spatial resource conservation while maintaining attack effectiveness. The framework enables practical deployment of comprehensive adversarial robustness testing in resource-constrained real-time applications where computational efficiency directly impacts mission success.

Individuals experiencing interpersonal violence (IPV), who depend on medical devices, represent a uniquely vulnerable population as healthcare technologies become increasingly connected. Despite rapid growth in MedTech innovation and "health-at-home" ecosystems, the intersection of MedTech cybersecurity and technology-facilitated abuse remains critically under-examined. IPV survivors who rely on therapeutic devices encounter a qualitatively different threat environment from the external, technically sophisticated adversaries typically modeled in MedTech cybersecurity research. We address this gap through two complementary methods: (1) the development of hazard-integrated threat models that fuse Cyber physical system security modeling with tech-abuse frameworks, and (2) an immersive simulation with practitioners, deploying a live version of our model, identifying gaps in digital forensic practice. Our hazard-integrated CIA threat models map exploits to acute and chronic biological effects, uncovering (i) Integrity attack pathways that facilitate "Medical gaslighting" and "Munchausen-by-IoMT", (ii) Availability attacks that create life-critical and sub-acute harms (glycaemic emergencies, blindness, mood destabilization), and (iii) Confidentiality threats arising from MedTech advertisements (geolocation tracking from BLE broadcasts). Our simulation demonstrates that these attack surfaces are unlikely to be detected in practice: participants overlooked MedTech, misclassified reproductive and assistive technologies, and lacked awareness of BLE broadcast artifacts. Our findings show that MedTech cybersecurity in IPV contexts requires integrated threat modeling and improved forensic capabilities for detecting, preserving and interpreting harms arising from compromised patient-technology ecosystems.

Virtual reality (VR) platforms and apps collect user sensor data, including motion, facial, eye, and hand data, in abstracted form. These data may expose users to unique privacy risks without their knowledge or meaningful awareness, yet the extent of these risks remains understudied. To address this gap, we propose VR ProfiLens, a framework to study user profiling based on VR sensor data and the resulting privacy risks across consumer VR apps. To systematically study this problem, we first develop a taxonomy rooted in the CCPA definition of personal information and expand it by sensor, app, and threat contexts to identify user attributes at risk. Then, we conduct a user study in which we collect VR sensor data from four sensor groups from real users interacting with 10 popular consumer VR apps, followed by a survey. We design and apply an analysis pipeline to demonstrate the feasibility of inferring user attributes using these data. Our results show that sensitive personal information can be inferred with moderately high to high risk (up to 90% F1 score) from abstracted sensor data. Through feature analysis, we further identify correlations among app groups and sensor groups in inferring user attributes. Our findings highlight risks to users, including privacy loss, tracking, targeted advertising, and safety threats. Finally, we discuss design implications and regulatory recommendations to enhance transparency and better protect users' privacy in VR.

The demand of customized large language models (LLMs) has led to commercial LLMs offering black-box fine-tuning APIs, yet this convenience introduces a critical security loophole: attackers could jailbreak the LLMs by fine-tuning them with malicious data. Though this security issue has recently been exposed, the feasibility of such attacks is questionable as malicious training dataset is believed to be detectable by moderation models such as Llama-Guard-3. In this paper, we propose TrojanPraise, a novel finetuning-based attack exploiting benign and thus filter-approved data. Basically, TrojanPraise fine-tunes the model to associate a crafted word (e.g., "bruaf") with harmless connotations, then uses this word to praise harmful concepts, subtly shifting the LLM from refusal to compliance. To explain the attack, we decouple the LLM's internal representation of a query into two dimensions of knowledge and attitude. We demonstrate that successful jailbreak requires shifting the attitude while avoiding knowledge shift, a distortion in the model's understanding of the concept. To validate this attack, we conduct experiments on five opensource LLMs and two commercial LLMs under strict black-box settings. Results show that TrojanPraise achieves a maximum attack success rate of 95.88% while evading moderation.

AI agents are autonomous systems that combine LLMs with external tools to solve complex tasks. While such tools extend capability, improper tool permissions introduce security risks such as indirect prompt injection and tool misuse. We characterize these failures as unbalanced tool-driven agency. Agents may retain unnecessary permissions (excessive agency) or fail to invoke required tools (insufficient agency), amplifying the attack surface and reducing performance. We introduce AgenTRIM, a framework for detecting and mitigating tool-driven agency risks without altering an agent's internal reasoning. AgenTRIM addresses these risks through complementary offline and online phases. Offline, AgenTRIM reconstructs and verifies the agent's tool interface from code and execution traces. At runtime, it enforces per-step least-privilege tool access through adaptive filtering and status-aware validation of tool calls. Evaluating on the AgentDojo benchmark, AgenTRIM substantially reduces attack success while maintaining high task performance. Additional experiments show robustness to description-based attacks and effective enforcement of explicit safety policies. Together, these results show that AgenTRIM provides a practical, capability-preserving approach to safer tool use in LLM-based agents.

Federated learning enables collaborative model training across distributed institutions without centralizing sensitive data; however, ensuring algorithmic fairness across heterogeneous data distributions while preserving privacy remains fundamentally unresolved. This paper introduces CryptoFair-FL, a novel cryptographic framework providing the first verifiable fairness guarantees for federated learning systems under formal security definitions. The proposed approach combines additively homomorphic encryption with secure multi-party computation to enable privacy-preserving verification of demographic parity and equalized odds metrics without revealing protected attribute distributions or individual predictions. A novel batched verification protocol reduces computational complexity from BigO(n^2) to BigO(n \log n) while maintaining (\dparam, \deltap)-differential privacy with dparam = 0.5 and deltap = 10^{-6}. Theoretical analysis establishes information-theoretic lower bounds on the privacy cost of fairness verification, demonstrating that the proposed protocol achieves near-optimal privacy-fairness tradeoffs. Comprehensive experiments across four benchmark datasets (MIMIC-IV healthcare records, Adult Income, CelebA, and a novel FedFair-100 benchmark) demonstrate that CryptoFair-FL reduces fairness violations from 0.231 to 0.031 demographic parity difference while incurring only 2.3 times computational overhead compared to standard federated averaging. The framework successfully defends against attribute inference attacks, maintaining adversarial success probability below 0.05 across all tested configurations. These results establish a practical pathway for deploying fairness-aware federated learning in regulated industries requiring both privacy protection and algorithmic accountability.

Cross-chain bridges constitute the single largest vector of systemic risk in Decentralized Finance (DeFi), accounting for over \$2.8 billion in losses since 2021. The fundamental vulnerability lies in the binary nature of existing bridge security models: a bridge is either fully operational or catastrophically compromised, with no intermediate state to contain partial failures. We present ASAS-BridgeAMM, a bridge-coupled automated market maker that introduces Contained Degradation: a formally specified operational state where the system gracefully degrades functionality in response to adversarial signals. By treating cross-chain message latency as a quantifiable execution risk, the protocol dynamically adjusts collateral haircuts, slippage bounds, and withdrawal limits. Across 18 months of historical replay on Ethereum and two auxiliary chains, ASAS-BridgeAMM reduces worst-case bridge-induced insolvency by 73% relative to baseline mint-and-burn architectures, while preserving 104.5% of transaction volume during stress periods. In rigorous adversarial simulations involving delayed finality, oracle manipulation, and liquidity griefing, the protocol maintains solvency with probability $>0.9999$ and bounds per-epoch bad debt to $<0.2%$ of total collateral. We provide a reference implementation in Solidity and formally prove safety (bounded debt), liveness (settlement completion), and manipulation resistance under a Byzantine relayer model.

Water distribution systems (WDSs) face increasing cyber-physical risks, which make reliable anomaly detection essential. Many data-driven models ignore network topology and are hard to interpret, while model-based ones depend strongly on parameter accuracy. This work proposes a hydraulic-aware graph attention network using normalized conservation law violations as features. It combines mass and energy balance residuals with graph attention and bidirectional LSTM to learn spatio-temporal patterns. A multi-scale module aggregates detection scores from node to network level. On the BATADAL dataset, it reaches $F1=0.979$, showing $3.3$pp gain and high robustness under $15\%$ parameter noise.

As LLMs rapidly advance and enter real-world use, their privacy implications are increasingly important. We study an authorship de-anonymization threat: using LLMs to link anonymous documents to their authors, potentially compromising settings such as double-blind peer review. We propose De-Anonymization at Scale (DAS), a large language model-based method for attributing authorship among tens of thousands of candidate texts. DAS uses a sequential progression strategy: it randomly partitions the candidate corpus into fixed-size groups, prompts an LLM to select the text most likely written by the same author as a query text, and iteratively re-queries the surviving candidates to produce a ranked top-k list. To make this practical at scale, DAS adds a dense-retrieval prefilter to shrink the search space and a majority-voting style aggregation over multiple independent runs to improve robustness and ranking precision. Experiments on anonymized review data show DAS can recover same-author texts from pools of tens of thousands with accuracy well above chance, demonstrating a realistic privacy risk for anonymous platforms. On standard authorship benchmarks (Enron emails and blog posts), DAS also improves both accuracy and scalability over prior approaches, highlighting a new LLM-enabled de-anonymization vulnerability.

Prompt injection attacks have become an increasing vulnerability for LLM applications, where adversarial prompts exploit indirect input channels such as emails or user-generated content to circumvent alignment safeguards and induce harmful or unintended outputs. Despite advances in alignment, even state-of-the-art LLMs remain broadly vulnerable to adversarial prompts, underscoring the urgent need for robust, productive, and generalizable detection mechanisms beyond inefficient, model-specific patches. In this work, we propose Zero-Shot Embedding Drift Detection (ZEDD), a lightweight, low-engineering-overhead framework that identifies both direct and indirect prompt injection attempts by quantifying semantic shifts in embedding space between benign and suspect inputs. ZEDD operates without requiring access to model internals, prior knowledge of attack types, or task-specific retraining, enabling efficient zero-shot deployment across diverse LLM architectures. Our method uses adversarial-clean prompt pairs and measures embedding drift via cosine similarity to capture subtle adversarial manipulations inherent to real-world injection attacks. To ensure robust evaluation, we assemble and re-annotate the comprehensive LLMail-Inject dataset spanning five injection categories derived from publicly available sources. Extensive experiments demonstrate that embedding drift is a robust and transferable signal, outperforming traditional methods in detection accuracy and operational efficiency. With greater than 93% accuracy in classifying prompt injections across model architectures like Llama 3, Qwen 2, and Mistral and a false positive rate of <3%, our approach offers a lightweight, scalable defense layer that integrates into existing LLM pipelines, addressing a critical gap in securing LLM-powered systems to withstand adaptive adversarial threats.

Large multimodal model powered GUI agents are emerging as high-privilege operators on mobile platforms, entrusted to perceive screen content and inject inputs across application boundaries. While these agents aim to automate complex tasks, we demonstrate that their design introduces a fundamental conflict with Android's strict application sandboxing. We present a novel cross-application Action Rebinding attack, which allows a malicious application with zero dangerous permissions to hijack the agent's execution and perform privileged operations on behalf of the attacker. Our attack exploits the inevitable observation-action gap inherent in the agent's reasoning pipeline. A malicious app can render a benign “contextual carrier” to elicit a planned action, and then swap the foreground to a sensitive target application during the reasoning latency. The agent, unaware of the transition, unwittingly executes the action in the privileged context. We further advance this attack by weaponizing the agent's own task-recovery logic to create programmable, multi-step exploit loops , and introducing an Intent Alignment Strategy (IAS) that manipulates the agent's reasoning to rationalize the hijacked state. We evaluate our attack on six widely-used Android GUI agents. Our results demonstrate a 100% success rate for atomic action hijacking and the ability to orchestrate high-impact exploits, including unauthorized file deletion, SMS transmission, and app uninstallation, without the attacker holding any corresponding permissions. Furthermore, since the malicious application separates intent from capability and contains no privileged API calls, it achieves a 0% detection rate across commercial malware scanners (e.g., VirusTotal), highlighting a critical blind spot in current mobile security analysis. To access experimental logs and demonstration videos, please contact yi_qian@smail.nju.edu.cn.

RAG has emerged as a key technique for enhancing response quality of LLMs without high computational cost. In traditional architectures, RAG services are provided by a single entity that hosts the dataset within a trusted local environment. However, individuals or small organizations often lack the resources to maintain data storage servers, leading them to rely on outsourced cloud storage. This dependence on untrusted third-party services introduces privacy risks. Embedding-based retrieval mechanisms, commonly used in RAG systems, are vulnerable to privacy leakage such as vector-to-text reconstruction attacks and structural leakage via vector analysis. Several privacy-preserving RAG techniques have been proposed but most existing approaches rely on partially homomorphic encryption, which incurs substantial computational overhead. To address these challenges, we propose an efficient privacy-preserving RAG framework (ppRAG) tailored for untrusted cloud environments that defends against vector-to-text attack, vector analysis, and query analysis. We propose Conditional Approximate Distance-Comparison-Preserving Symmetric Encryption (CAPRISE) that encrypts embeddings while still allowing the cloud to compute similarity between an encrypted query and the encrypted database embeddings. CAPRISE preserves only the relative distance ordering between the encrypted query and each encrypted database embedding, without exposing inter-database distances, thereby enhancing both privacy and efficiency. To mitigate query analysis, we introduce DP by perturbing the query embedding prior to encryption, preventing the cloud from inferring sensitive patterns. Experimental results show that ppRAG achieves efficient processing throughput, high retrieval accuracy, strong privacy guarantees, making it a practical solution for resource-constrained users seeking secure cloud-augmented LLMs.

The emergence of 6G-enabled vehicular metaverses enables Autonomous Vehicles (AVs) to operate across physical and virtual spaces through space-air-ground-sea integrated networks. The AVs can deploy AI agents powered by large AI models as personalized assistants, on edge servers to support intelligent driving decision making and enhanced on-board experiences. However, such cross-reality interactions may cause serious location privacy risks, as adversaries can infer AV trajectories by correlating the location reported when AVs request LBS in reality with the location of the edge servers on which their corresponding AI agents are deployed in virtuality. To address this challenge, we design a cross-reality location privacy protection framework based on hybrid actions, including continuous location perturbation in reality and discrete privacy-aware AI agent migration in virtuality. In this framework, a new privacy metric, termed cross-reality location entropy, is proposed to effectively quantify the privacy levels of AVs. Based on this metric, we formulate an optimization problem to optimize the hybrid action, focusing on achieving a balance between location protection, service latency reduction, and quality of service maintenance. To solve the complex mixed-integer problem, we develop a novel LLM-enhanced Hybrid Diffusion Proximal Policy Optimization (LHDPPO) algorithm, which integrates LLM-driven informative reward design to enhance environment understanding with double Generative Diffusion Models-based policy exploration to handle high-dimensional action spaces, thereby enabling reliable determination of optimal hybrid actions. Extensive experiments on real-world datasets demonstrate that the proposed framework effectively mitigates cross-reality location privacy leakage for AVs while maintaining strong user immersion within 6G-enabled vehicular metaverse scenarios.

The increasing prevalence of Large Language Models (LLMs) demands effective safeguards for their operation, particularly concerning their tendency to generate out-of-context responses. A key challenge is accurately detecting when LLMs stray from expected conversational norms, manifesting as topic shifts, factual inaccuracies, or outright hallucinations. Traditional anomaly detection struggles to directly apply within contextual semantics. This paper outlines our experiment in exploring the use of Representation Engineering (RepE) and One-Class Support Vector Machine (OCSVM) to identify subspaces within the internal states of LLMs that represent a specific context. By training OCSVM on in-context examples, we establish a robust boundary within the LLM's hidden state latent space. We evaluate out study with two open source LLMs - Llama and Qwen models in specific contextual domain. Our approach entailed identifying the optimal layers within the LLM's internal state subspaces that strongly associates with the context of interest. Our evaluation results showed promising results in identifying the subspace for a specific context. Aside from being useful in detecting in or out of context conversation threads, this research work contributes to the study of better interpreting LLMs.

The rapid evolution of the automobile sector, driven by advancements in connected and autonomous vehicles (CAVs), has transformed how vehicles communicate, operate, and interact with their surroundings. Technologies such as Vehicle-to-Everything (V2X) communication enable autonomous cars to generate and exchange substantial amounts of data with real-world entities, enhancing safety, improving performance, and delivering personalized user experiences. However, this data-driven ecosystem introduces significant challenges, particularly concerning data privacy, security, and governance. The absence of transparency and comprehensive regulatory frameworks exacerbates issues of unauthorized data access, prolonged retention, and potential misuse, creating tension between consumer benefits and privacy risks. This review paper explores the multifaceted nature of data sharing in CAVs, analyzing its contributions to innovation and its associated vulnerabilities. It evaluates data-sharing mechanisms and communication technologies, highlights the benefits of data exchange across various use cases, examines privacy concerns and risks of data misuse, and critically reviews regulatory frameworks and their inadequacies in safeguarding user privacy. By providing a thorough analysis of the current state of data sharing in the automotive sector, the paper emphasizes the urgent need for robust policies and ethical data management practices. It calls for striking a balance between fostering technological advancements and ensuring secure, consumer-friendly solutions, paving the way for a trustworthy and innovative automotive future.

Traditional side-channels take advantage of secrets being used as inputs to unsafe instructions, used for memory accesses, or used in control flow decisions. Constant-time programming, which restricts such code patterns, has been widely adopted as a defense against these vulnerabilities. However, new hardware optimizations in the form of Data Memory-dependent Prefetchers (DMP) present in Apple, Intel, and ARM CPUs have shown such defenses are not sufficient. These prefetchers, unlike classical prefetchers, use the content of memory as well as the trace of prior accesses to determine prefetch targets. An adversary abusing such a prefetcher has been shown to be able to mount attacks leaking data-at-rest; data that is never used by the program, even speculatively, in an unsafe manner. In response, this paper introduces SplittingSecrets, a compiler-based tool that can harden software libraries against side-channels arising from DMPs. SplittingSecrets's approach avoids reasoning about the complex internals of different DMPs and instead relies on one key aspect of all DMPs: activation requires data to resemble addresses. To prevent secret data from leaking, SplittingSecrets transforms memory operations to ensure that secrets are never stored in memory in a manner resembling an address, thereby avoiding DMP activation on those secrets. Rather than disable a DMP entirely, SplittingSecrets can provide targeted hardening for only specific secrets entirely in software. We have implemented SplittingSecrets using LLVM, supporting both source-level memory operations and those generated by the compiler backend for the AArch64 architecture, We have analyzed the performance overhead involved in safeguarding secrets from DMP-induced attacks using common primitives in libsodium, a popular cryptographic library when built for Apple M-series CPUs.

Insider threat detection is a key challenge in enterprise security, relying on user activity logs that capture rich and complex behavioral patterns. These logs are often multi-channel, non-stationary, and anomalies are rare, making anomaly detection challenging. To address these issues, we propose a novel framework that integrates wavelet-aware modulation, multi-resolution wavelet decomposition, and resolution-adaptive attention for robust anomaly detection. Our approach first applies a deviation-aware modulation scheme to suppress routine behaviors while amplifying anomalous deviations. Next, discrete wavelet transform (DWT) decomposes the log signals into multi-resolution representations, capturing both long-term trends and short-term anomalies. Finally, a learnable attention mechanism dynamically reweights the most discriminative frequency bands for detection. On the CERT r4.2 benchmark, our approach consistently outperforms existing baselines in precision, recall, and F1 score across various time granularities and scenarios.

Hard-label black-box attacks, relying solely on top-1 predictions, represent one of the most challenging yet practically threat models. Despite recent progress, existing approaches face two key limitations: (1) they overlook the critical role of initialization, focusing primarily on optimization strategies; and (2) they rely heavily on empirical heuristics without theoretical guarantees. To bridge this gap, we establish a unified theoretical framework showing that existing sign-flipping hard-label attacks can be understood as approximating the true gradient sign. Guided by this principled analysis, we propose a novel attack framework featuring a zero-query initialization strategy and a Pattern-Driven Optimization (PDO) algorithm. We provide theoretical guarantees that our initialization yields higher cosine similarity to the true gradient sign than random baselines, and our PDO module achieves significantly lower query complexity than baseline search methods. Extensive experiments across CIFAR-10, ImageNet, and ObjectNet-covering standard and adversarially trained models, commercial APIs, and CLIP models-demonstrate that our method consistently outperforms SOTA hard-label attacks in both success rate and efficiency, particularly under low query budgets. Furthermore, our method demonstrates robust generalization across corrupted data (ImageNet-C), biomedical images (PathMNIST), and dense prediction tasks such as segmentation. Notably, it bypasses the stateful defense Blacklight, achieving a 0% detection rate.

Analysis of clinical data is a cornerstone of biomedical research with applications in areas such as genomic testing and response characterization of therapeutic drugs. Maintaining strict privacy controls is essential because such data typically contains personally identifiable health information of patients. At the same time, regulatory compliance often requires study managers to demonstrate the integrity and authenticity of participant data used in analyses. Balancing these competing requirements of privacy preservation and verifiable accountability remains a critical challenge. In this paper, we present CoSMeTIC, a zero-knowledge computational framework that proposes computational Sparse Merkle Trees (SMTs) as a means to generate verifiable inclusion and exclusion proofs for individual participants' data in clinical studies. We formally analyze the zero-knowledge properties of CoSMeTIC and evaluate its computational efficiency through extensive experiments. We demonstrate the framework on Huntington's disease and HIV-1 case studies, using simulated CAG-repeat cohorts derived from published summary statistics and published de-identified clinical lab measurements of virus samples. Using two-sample Kolmogorov-Smirnov and likelihood-ratio hypothesis tests, along with logistic-regression-based genomic analyses on the de-identified datasets, we show that CoSMeTIC achieves strong privacy guarantees while maintaining statistical fidelity. Our results suggest that CoSMeTIC provides a scalable and practical alternative for achieving regulatory compliance with rigorous privacy protection in large-scale clinical research.

Personalized health analytics increasingly rely on population benchmarks to provide contextual insights such as ”How do I compare to others like me?” However, cohort-based aggregation of health data introduces nontrivial privacy risks, particularly in interactive and longitudinal digital platforms. Existing privacy frameworks such as $k$-anonymity and differential privacy provide essential but largely static guarantees that do not fully capture the cumulative, distributional, and tail-dominated nature of re-identification risk in deployed systems. In this work, we present a privacy-preserving cohort analytics framework that combines deterministic cohort constraints, differential privacy mechanisms, and synthetic baseline generation to enable personalized population comparisons while maintaining strong privacy protections. We further introduce a stochastic risk modeling approach that treats re-identification risk as a random variable evolving over time, enabling distributional evaluation through Monte Carlo simulation. Adapting quantitative risk measures from financial mathematics, we define Privacy Loss at Risk (P-VaR) to characterize worst-case privacy outcomes under realistic cohort dynamics and adversary assumptions. We validate our framework through system-level analysis and simulation experiments, demonstrating how privacy-utility tradeoffs can be operationalized for digital health platforms. Our results suggest that stochastic risk modeling complements formal privacy guarantees by providing interpretable, decision-relevant metrics for platform designers, regulators, and clinical informatics stakeholders.

Fine-tuned language models pose significant privacy risks, as they may memorize and expose sensitive information from their training data. Membership inference attacks (MIAs) provide a principled framework for auditing these risks, yet existing methods achieve limited detection rates, particularly at the low false-positive thresholds required for practical privacy auditing. We present EZ-MIA, a membership inference attack that exploits a key observation: memorization manifests most strongly at error positions, specifically tokens where the model predicts incorrectly yet still shows elevated probability for training examples. We introduce the Error Zone (EZ) score, which measures the directional imbalance of probability shifts at error positions relative to a pretrained reference model. This principled statistic requires only two forward passes per query and no model training of any kind. On WikiText with GPT-2, EZ-MIA achieves 3.8x higher detection than the previous state-of-the-art under identical conditions (66.3% versus 17.5% true positive rate at 1% false positive rate), with near-perfect discrimination (AUC 0.98). At the stringent 0.1% FPR threshold critical for real-world auditing, we achieve 8x higher detection than prior work (14.0% versus 1.8%), requiring no reference model training. These gains extend to larger architectures: on AG News with Llama-2-7B, we achieve 3x higher detection (46.7% versus 15.8% TPR at 1% FPR). These results establish that privacy risks of fine-tuned language models are substantially greater than previously understood, with implications for both privacy auditing and deployment decisions. Code is available at https://github.com/JetBrains-Research/ez-mia.