Papers

As the paradigm of Human-Centered AI (HCAI) gains prominence, its benefits to society are accompanied by significant ethical concerns, one of which is the protection of individual privacy. This chapter provides a comprehensive overview of privacy within HCAI, proposing a human-centered privacy (HCP) framework, providing integrated solution from technology, ethics, and human factors perspectives. The chapter begins by mapping privacy risks across each stage of AI development lifecycle, from data collection to deployment and reuse, highlighting the impact of privacy risks on the entire system. The chapter then introduces privacy-preserving techniques such as federated learning and dif erential privacy. Subsequent chapters integrate the crucial user perspective by examining mental models, alongside the evolving regulatory and ethical landscapes as well as privacy governance. Next, advice on design guidelines is provided based on the human-centered privacy framework. After that, we introduce practical case studies across diverse fields. Finally, the chapter discusses persistent open challenges and future research directions, concluding that a multidisciplinary approach, merging technical, design, policy, and ethical expertise, is essential to successfully embed privacy into the core of HCAI, thereby ensuring these technologies advance in a manner that respects and ensures human autonomy, trust and dignity.

We prove the conjecture stated in Appendix F.3 of \citet{zhu2022optimalaccountingdifferentialprivacy}: among all conversion rules that map a Rényi Differential Privacy (RDP) profile $\tau \mapsto \rho(\tau)$ to a valid hypothesis-testing trade-off $f$, the rule based on the intersection of single-order RDP privacy regions is optimal. This optimality holds simultaneously for all valid RDP profiles and for all Type I error levels $\alpha$. Concretely, we show that in the space of trade-off functions, the tightest possible bound is $f_{\rho(\cdot)}(\alpha) = \sup_{\tau \geq 0.5} f_{\tau,\rho(\tau)}(\alpha)$: the pointwise maximum of the single-order bounds for each RDP privacy region. Our proof unifies and sharpens the insights of \citet{balle2019hypothesistestinginterpretationsrenyi}, \citet{asoodeh2021variantsdifferentialprivacylossless}, and \citet{zhu2022optimalaccountingdifferentialprivacy}. Our analysis relies on a precise geometric characterization of the RDP privacy region, leveraging its convexity and the fact that its boundary is determined exclusively by Bernoulli mechanisms. Our results establish that the \enquote{intersection-of-RDP-privacy-regions} rule is not only valid, but optimal: no other black-box conversion can uniformly dominate it in the Blackwell sense, marking the fundamental limit of what can be inferred about a mechanism's privacy solely from its RDP guarantees.

Mixture-of-Experts (MoE) language models introduce unique challenges for safety alignment due to their sparse routing mechanisms, which can enable degenerate optimization behaviors under standard full-parameter fine-tuning. In our preliminary experiments, we observe that naively applying full-parameter safety fine-tuning to MoE models can reduce attack success rates through routing or expert dominance effects, rather than by directly repairing Safety-Critical Experts. To address this challenge, we propose RASA, a routing-aware expert-level alignment framework that explicitly repairs Safety-Critical Experts while preventing routing-based bypasses. RASA identifies experts disproportionately activated by successful jailbreaks, selectively fine-tunes only these experts under fixed routing, and subsequently enforces routing consistency with safety-aligned contexts. Across two representative MoE architectures and a diverse set of jailbreak attacks, RASA achieves near-perfect robustness, strong cross-attack generalization, and substantially reduced over-refusal, while preserving general capabilities on benchmarks such as MMLU, GSM8K, and TruthfulQA. Our results suggest that robust MoE safety alignment benefits from targeted expert repair rather than global parameter updates, offering a practical and architecture-preserving alternative to prior approaches.

Cryptographic operations are critical for securing IoT, edge computing, and autonomous systems. However, current RISC-V platforms lack efficient hardware support for comprehensive cryptographic algorithm families and post-quantum cryptography. This paper presents Crypto-RV, a RISC-V co-processor architecture that unifies support for SHA-256, SHA-512, SM3, SHA3-256, SHAKE-128, SHAKE-256 AES-128, HARAKA-256, and HARAKA-512 within a single 64-bit datapath. Crypto-RV introduces three key architectural innovations: a high-bandwidth internal buffer (128x64-bit), cryptography-specialized execution units with four-stage pipelined datapaths, and a double-buffering mechanism with adaptive scheduling optimized for large-hash. Implemented on Xilinx ZCU102 FPGA at 160 MHz with 0.851 W dynamic power, Crypto-RV achieves 165 times to 1,061 times speedup over baseline RISC-V cores, 5.8 times to 17.4 times better energy efficiency compared to powerful CPUs. The design occupies only 34,704 LUTs, 37,329 FFs, and 22 BRAMs demonstrating viability for high-performance, energy-efficient cryptographic processing in resource-constrained IoT environments.

Effective demand forecasting is crucial for reducing food waste. However, data privacy concerns often hinder collaboration among retailers, limiting the potential for improved predictive accuracy. In this study, we explore the application of Federated Learning (FL) in Sustainable Supply Chain Management (SSCM), with a focus on the grocery retail sector dealing with perishable goods. We develop a baseline predictive model for demand forecasting and waste assessment in an isolated retailer scenario. Subsequently, we introduce a Blockchain-based FL model, trained collaboratively across multiple retailers without direct data sharing. Our preliminary results show that FL models have performance almost equivalent to the ideal setting in which parties share data with each other, and are notably superior to models built by individual parties without sharing data, cutting waste and boosting efficiency.

Large Language Models (LLMs) face increasing threats from jailbreak attacks that bypass safety alignment. While prompt-based defenses such as Role-Oriented Prompts (RoP) and Task-Oriented Prompts (ToP) have shown effectiveness, the role of few-shot demonstrations in these defense strategies remains unclear. Prior work suggests that few-shot examples may compromise safety, but lacks investigation into how few-shot interacts with different system prompt strategies. In this paper, we conduct a comprehensive evaluation on multiple mainstream LLMs across four safety benchmarks (AdvBench, HarmBench, SG-Bench, XSTest) using six jailbreak attack methods. Our key finding reveals that few-shot demonstrations produce opposite effects on RoP and ToP: few-shot enhances RoP's safety rate by up to 4.5% through reinforcing role identity, while it degrades ToP's effectiveness by up to 21.2% through distracting attention from task instructions. Based on these findings, we provide practical recommendations for deploying prompt-based defenses in real-world LLM applications.

Cloud-native application platforms and latency-sensitive systems such as 5G Core networks rely heavily on certificate-based Public Key Infrastructure (PKI) and mutual TLS to secure service-to-service communication. While effective, this model introduces significant operational and performance overhead, which is further amplified in the post-quantum setting due to large certificates and expensive signature verification. In this paper, we present a certificate-free authentication framework for private distributed systems based on post-quantum Identity-Based Encryption(IBE). Our design replaces certificate and signature based authentication with identity-derived keys and identity-based key encapsulation, enabling mutually authenticated TLS connections without certificate transmission or validation. We describe an IBE-based replacement for private PKI, including identity lifecycle management, and show how it can be instantiated using a threshold Private Key Generator (T-PKG). We apply this framework to cloud-native application deployments and latency-sensitive 5G Core networks. In particular, we demonstrate how identity-based TLS integrates with the 5G Service-Based Architecture while preserving security semantics and 3GPP requirements, and we show how the same architecture can replace private PKI in Kubernetes, including its control plane, without disrupting existing trust domains or deployment models.

Large Reasoning Models (LRMs) have achieved tremendous success with their chain-of-thought (CoT) reasoning, yet also face safety issues similar to those of basic language models. In particular, while algorithms are designed to guide them to deliberately refuse harmful prompts with safe reasoning, this process often fails to generalize against diverse and complex jailbreak attacks. In this work, we attribute these failures to the generalization of the safe reasoning process, particularly their insufficiency against complex attack prompts. We provide both theoretical and empirical evidence to show the necessity of a more sufficient safe reasoning process to defend against advanced attack prompts. Building on this insight, we propose a Risk-Aware Preference Optimization (RAPO) framework that enables LRM to adaptively identify and address the safety risks with appropriate granularity in its thinking content. Extensive experiments demonstrate that RAPO successfully generalizes multiple LRMs' safe reasoning adaptively across diverse attack prompts whilst preserving general utility, contributing a robust alignment technique for LRM safety. Our code is available at https://github.com/weizeming/RAPO.

Denial-of-Service (DoS) conditions in enterprise networks are commonly attributed to malicious actors. However, availability can also be compromised by benign non-malicious insider behavior. This paper presents an empirical study of a production enterprise LAN that demonstrates how routine docking and undocking of user endpoints repeatedly trigger rapid recalculations of the control plane of the Rapid Spanning Tree Protocol (RSTP) [1]. Although protocol-compliant and nonmalicious, these events introduce transient forwarding disruptions of approximately 2-4 seconds duration that degrade realtime streaming (voice and video) services while remaining largely undetected by conventional security monitoring. We map this phenomenon to the NIST and MITRE insider threat frameworks, characterizing it as an unintentional insider-driven availability breach, and demonstrate that explicit edge-port configuration effectively mitigates the condition without compromising loop prevention

Large language models (LLMs) for Verilog code generation are increasingly adopted in hardware design, yet remain vulnerable to backdoor attacks where adversaries inject malicious triggers during training to induce vulnerable hardware designs. Unlike patchable software vulnerabilities, hardware trojans become irreversible once fabricated, making remediation extremely costly or impossible. Existing active defenses require access to training data, impractical for third-party LLM users, while passive defenses struggle against semantically stealthy triggers that naturally blend into design specifications. In this paper, we hypothesize that under the requirements of both effectiveness and stealthiness, attackers are strongly biased toward embedding triggers in non-functional requirements (e.g., style modifiers, quality descriptors) rather than functional specifications that determine hardware behavior. Exploiting this insight, we propose Semantic Consensus Decoding (SCD), an inference-time passive defense with two key components: (1) functional requirement extraction that identifies essential requirements from user specifications, and (2) consensus decoding that adaptively fuses output distributions based on full user specifications and extracted functional requirements. When these distributions diverge significantly, SCD automatically suppresses suspicious components. Extensive experiments with three representative backdoor attacks demonstrate that SCD reduces average attack success rate from 89% to under 3% with negligible impact on generation quality.

Gradient boosted decision trees, particularly XGBoost, are among the most effective methods for tabular data. As deployment in sensitive settings increases, cryptographic guarantees of model integrity become essential. We present ZKBoost, the first zero-knowledge proof of training (zkPoT) protocol for XGBoost, enabling model owners to prove correct training on a committed dataset without revealing data or model parameters. Naively re-executing XGBoost training in ZK would incur prohibitive costs, primarily due to the oblivious partitioning of training samples and unknown tree splits. Moreover, previous work on ZKP of training and inference had subtle security issues, such as leakage of tree topology and soundness gaps allowing cheating model providers to deviate from the correct execution of training and inference. We make two key contributions to address these challenges: (1) a generic zkPoT template for XGBoost that can be instantiated with any general-purpose ZKP backend, significantly improving prover costs compared to naive re-execution of the training process; and (2) a VOLE-based instantiation that overcomes the security issues of previous ZK proofs of training at minimal costs. To maximize efficiency, we develop a fixed-point version of XGBoost, which is particularly well suited for efficient instantiation of ZKP, and show it matches standard XGBoost accuracy to within 1% on real-world datasets.

We present a text-reconstruction attack on mixture-of-experts (MoE) language models that recovers tokens from expert selections alone. In MoE models, each token is routed to a subset of expert subnetworks; we show these routing decisions leak substantially more information than previously understood. Prior work using logistic regression achieves limited reconstruction; we show that a 3-layer MLP improves this to 63.1% top-1 accuracy, and that a transformer-based sequence decoder recovers 91.2% of tokens top-1 (94.8% top-10) on 32-token sequences from OpenWebText after training on 100M tokens. These results connect MoE routing to the broader literature on embedding inversion. We outline practical leakage scenarios (e.g., distributed inference and side channels) and show that adding noise reduces but does not eliminate reconstruction. Our findings suggest that expert selections in MoE deployments should be treated as sensitive as the underlying text.

Reproducible container builds promise a simple integrity check for software supply chains: rebuild an image from its Dockerfile and compare hashes. We build a Docker measurement pipeline and apply it to a stratified sample of 2,000 GitHub repositories that contained a Dockerfile. We found that only 56% produce any buildable image, and just 2.7% of those are bitwise reproducible without any infrastructure configurations. After modifying infrastructure configurations, we raise bitwise reproducibility by 18.6%, but 78.7% of buildable Dockerfiles remain non-reproducible. We analyze the root causes of the remaining differences, and find that beyond timestamps and metadata, developer-controlled choices such as uncleaned caches, logs, documentation, and floating versions are dominant causes of non-reproducibility. We derive concrete Dockerfile guidelines from these patterns and discuss how they can inform future linters and Continuous Integration (CI) checks for reproducible containers.

We present a data poisoning attack – Phantom Transfer – with the property that, even if you know precisely how the poison was placed into an otherwise benign dataset, you cannot filter it out. We achieve this by modifying subliminal learning to work in real-world contexts and demonstrate that the attack works regardless of which model produced the data, which model is trained on the data or what the attack target is. Furthermore, the attack survives 11 tested data-level defences, including one where every sample is paraphrased by another model. We characterise when this attack works best and show that it can be used to plant password-triggered behaviours into models while still beating defences. We suggest that future defences should be supplemented with white-box methods and post-training model audits.

Text-to-image (T2I) diffusion models are widely adopted for their strong generative capabilities, yet remain vulnerable to backdoor attacks. Existing attacks typically rely on fixed textual triggers and single-entity backdoor targets, making them highly susceptible to enumeration-based input defenses and attention-consistency detection. In this work, we propose Semantic-level Backdoor Attack (SemBD), which introduces representation-level triggers based on continuous semantic regions rather than discrete textual patterns. SemBD implants such semantic backdoors by distillation-based editing of the key and value projection matrices in cross-attention layers, enabling semantically equivalent but textually diverse prompts to activate the backdoor. To further enhance stealthiness, SemBD incorporates a semantic regularization to prevent unintended activation under incomplete semantics, as well as multi-entity backdoor targets that avoid highly consistent cross-attention patterns. Extensive experiments demonstrate that SemBD achieves a 100% attack success rate while maintaining strong robustness against state-of-the-art input-level defenses. Our code is available at https://github.com/DPAS-Lab/SemBD/.

Activation steering is a practical post-training model alignment technique to enhance the utility of Large Language Models (LLMs). Prior to deploying a model as a service, developers can steer a pre-trained model toward specific behavioral objectives, such as compliance or instruction adherence, without the need for retraining. This process is as simple as adding a steering vector to the model's internal representations. However, this capability unintentionally introduces critical and under-explored safety risks. We identify a phenomenon termed Steering Externalities, where steering vectors derived from entirely benign datasets-such as those enforcing strict compliance or specific output formats like JSON-inadvertently erode safety guardrails. Experiments reveal that these interventions act as a force multiplier, creating new vulnerabilities to jailbreaks and increasing attack success rates to over 80% on standard benchmarks by bypassing the initial safety alignment. Ultimately, our results expose a critical blind spot in deployment: benign activation steering systematically erodes the "safety margin," rendering models more vulnerable to black-box attacks and proving that inference-time utility improvements must be rigorously audited for unintended safety externalities.

We study privacy amplification by synthetic data release, a phenomenon in which differential privacy guarantees are improved by releasing only synthetic data rather than the private generative model itself. Recent work by Pierquin et al. (2025) established the first formal amplification guarantees for a linear generator, but they apply only in asymptotic regimes where the model dimension far exceeds the number of released synthetic records, limiting their practical relevance. In this work, we show a surprising result: under a bounded-parameter assumption, privacy amplification persists even when releasing an unbounded number of synthetic records, thereby improving upon the bounds of Pierquin et al. (2025). Our analysis provides structural insights that may guide the development of tighter privacy guarantees for more complex release mechanisms.

Large language models (LLMs) have been widely adopted in modern software development lifecycles, where they are increasingly used to automate and assist code generation, significantly improving developer productivity and reducing development time. In the blockchain domain, developers increasingly rely on LLMs to generate and maintain smart contracts, the immutable, self-executing components of decentralized applications. Because deployed smart contracts cannot be modified, correctness and security are paramount, particularly in high-stakes domains such as finance and governance. Despite this growing reliance, the security implications of LLM-generated smart contracts remain insufficiently understood. In this work, we conduct a systematic security analysis of Solidity smart contracts generated by state-of-the-art LLMs, including ChatGPT, Gemini, and Sonnet. We evaluate these contracts against a broad set of known smart contract vulnerabilities to assess their suitability for direct deployment in production environments. Our extensive experimental study shows that, despite their syntactic correctness and functional completeness, LLM-generated smart contracts frequently exhibit severe security flaws that could be exploited in real-world settings. We further analyze and categorize these vulnerabilities, identifying recurring weakness patterns across different models. Finally, we discuss practical countermeasures and development guidelines to help mitigate these risks, offering actionable insights for both developers and researchers. Our findings aim to support safe integration of LLMs into smart contract development workflows and to strengthen the overall security of the blockchain ecosystem against future security failures.

This work presents a consensus-based Bayesian framework to detect malicious user behavior in enterprise directory access graphs. By modeling directories as topics and users as agents within a multi-level interaction graph, we simulate access evolution using influence-weighted opinion dynamics. Logical dependencies between users are encoded in dynamic matrices Ci, and directory similarity is captured via a shared influence matrix W. Malicious behavior is injected as cross-component logical perturbations that violate structural norms of strongly connected components(SCCs). We apply theoretical guarantees from opinion dynamics literature to determine topic convergence and detect anomaly via scaled opinion variance. To quantify uncertainty, we introduce a Bayesian anomaly scoring mechanism that evolves over time, using both static and online priors. Simulations over synthetic access graphs validate our method, demonstrating its sensitivity to logical inconsistencies and robustness under dynamic perturbation.

Block space on the blockchain is scarce and must be allocated efficiently through block building. However, Ethereum's current block-building ecosystem, MEV-Boost, has become highly centralized due to integration, which distorts competition, reduces blockspace efficiency, and obscures MEV flow transparency. To guarantee equitability and economic efficiency in block building, we propose $\mathrm{Boost+}$, a system that decouples the process into collecting and ordering transactions, and ensures equal access to all collected transactions. The core of $\mathrm{Boost+}$ is the mechanism $\mathit{M}_{\mathrm{Boost+}}$, built around a default algorithm. $\mathit{M}_{\mathrm{Boost+}}$ aligns incentives for both searchers (intermediaries that generate or route transactions) and builders: Truthful bidding is a dominant strategy for all builders. For searchers, truthful reporting is dominant whenever the default algorithm dominates competing builders, and it remains dominant for all conflict-free transactions, even when builders may win. We further show that even if a searcher can technically integrate with a builder, non-integration combined with truthful bidding still dominates any deviation for conflict-free transactions. We also implement a concrete default algorithm informed by empirical analysis of real-world transactions and evaluate its efficacy using historical transaction data.

In sensitive applications involving relational datasets, protecting information about individual links from adversarial queries is of paramount importance. In many such settings, the available data are summarized solely through the degrees of the nodes in the network. We adopt the $\beta$ model, which is the prototypical statistical model adopted for this form of aggregated relational information, and study the problem of minimax-optimal parameter estimation under both local and central differential privacy constraints. We establish finite sample minimax lower bounds that characterize the precise dependence of the estimation risk on the network size and the privacy parameters, and we propose simple estimators that achieve these bounds up to constants and logarithmic factors under both local and central differential privacy frameworks. Our results provide the first comprehensive finite sample characterization of privacy utility trade offs for parameter estimation in $\beta$ models, addressing the classical graph case and extending the analysis to higher order hypergraph models. We further demonstrate the effectiveness of our methods through experiments on synthetic data and a real world communication network.

Prompt injection attacks manipulate webpage content to cause web agents to execute attacker-specified tasks instead of the user's intended ones. Existing methods for detecting and localizing such attacks achieve limited effectiveness, as their underlying assumptions often do not hold in the web-agent setting. In this work, we propose WebSentinel, a two-step approach for detecting and localizing prompt injection attacks in webpages. Given a webpage, Step I extracts segments of interest that may be contaminated, and Step II evaluates each segment by checking its consistency with the webpage content as context. We show that WebSentinel is highly effective, substantially outperforming baseline methods across multiple datasets of both contaminated and clean webpages that we collected. Our code is available at: https://github.com/wxl-lxw/WebSentinel.

Everyday services of society increasingly rely on mobile applications, resulting in a conflicting situation between the possibility of participation on the one side and user privacy and digital freedom on the other. In order to protect users' rights to informational self-determination, regulatory approaches for the collection and processing of personal data have been developed, such as the EU's GDPR. However, inspecting the compliance of mobile apps with privacy regulations remains difficult. Thus, in order to enable end users and enforcement bodies to verify and enforce data protection compliance, we propose mopri, a conceptual framework designed for analyzing the behavior of mobile apps through a comprehensive, adaptable, and user-centered approach. Recognizing the gaps in existing frameworks, mopri serves as a foundation for integrating various analysis tools into a streamlined, modular pipeline that employs static and dynamic analysis methods. Building on this concept, a prototype has been developed which effectively extracts permissions and tracking libraries while employing robust methods for dynamic traffic recording and decryption. Additionally, it incorporates result enrichment and reporting features that enhance the clarity and usability of the analysis outcomes. The prototype showcases the feasibility of a holistic and modular approach to privacy analysis, emphasizing the importance of continuous adaptation to the evolving challenges presented by the mobile app ecosystem.

Hardware Trojans (HTs) threaten the trust and reliability of integrated circuits (ICs), particularly when triggered HTs remain dormant during standard testing and activate only under rare conditions. Existing electromagnetic (EM) side-channel-based detection techniques often rely on golden references or labeled data, which are infeasible in modern distributed manufacturing. This paper introduces a reference-free, design-agnostic framework for detecting triggered HTs directly from post-silicon EM emissions. The proposed flow converts each EM trace into a time-frequency scalogram using Continuous Wavelet Transform (CWT), extracts discriminative features through a convolutional neural network (CNN), reduces dimensionality with principal component analysis (PCA), and applies Bayesian Gaussian Mixture Modeling (BGMM) for unsupervised probabilistic clustering. The framework quantifies detection confidence using posterior-based metrics (alpha_{post}, beta_{post}), Bayesian information criterion (Delta BIC), and Mahalanobis cluster separation (D), enabling interpretable anomaly decisions without golden data. Experimental validation on AES-128 designs embedded with four different HTs demonstrates high separability between HT-free and HT-activated conditions and robustness to PCA variance thresholds. The results highlight the method's scalability, statistical interpretability, and potential for extension to runtime and in-field HT monitoring in trusted microelectronics.

We investigated the capabilities of GPT-4o and Gemini 2.0 Flash for secure Infrastructure as Code (IaC) development. For security smell detection, on the Stack Overflow dataset, which primarily contains small, simplified code snippets, the models detected at least 71% of security smells when prompted to analyze code from a security perspective (general prompt). With a guided prompt (adding clear, step-by-step instructions), this increased to 78%.In GitHub repositories, which contain complete, real-world project scripts, a general prompt was less effective, leaving more than half of the smells undetected. However, with the guided prompt, the models uncovered at least 67% of the smells. For secure code generation, we prompted LLMs with 89 vulnerable synthetic scenarios and observed that only 7% of the generated scripts were secure. Adding an explicit instruction to generate secure code increased GPT secure output rate to 17%, while Gemini changed little (8%). These results highlight the need for further research to improve LLMs' capabilities in assisting developers with secure IaC development.

The Model Context Protocol (MCP) enables large language models to invoke external tools through natural-language descriptions, forming the foundation of many AI agent applications. However, MCP does not enforce consistency between documented tool behavior and actual code execution, even though MCP Servers often run with broad system privileges. This gap introduces a largely unexplored security risk. We study how mismatches between externally presented tool descriptions and underlying implementations systematically shape the mental models and decision-making behavior of intelligent agents. Specifically, we present the first large-scale study of description-code inconsistency in the MCP ecosystem. We design an automated static analysis framework and apply it to 10,240 real-world MCP Servers across 36 categories. Our results show that while most servers are highly consistent, approximately 13% exhibit substantial mismatches that can enable undocumented privileged operations, hidden state mutations, or unauthorized financial actions. We further observe systematic differences across application categories, popularity levels, and MCP marketplaces. Our findings demonstrate that description-code inconsistency is a concrete and prevalent attack surface in MCP-based AI agents, and motivate the need for systematic auditing and stronger transparency guarantees in future agent ecosystems.

Decentralized Pliable Index Coding (DPIC) problem addresses efficient information exchange in distributed systems where clients communicate among themselves without a central server. An important consideration in DPIC is the heterogeneity of side-information and demand sizes. Although many prior works assume homogeneous settings with identical side-information cardinality and single message demands, these assumptions limit real-world applicability where clients typically possess unequal amounts of prior information. In this paper, we study DPIC problem under heterogeneous side-information cardinalities. We propose a transmission scheme that coordinates client broadcasts to maximize coding efficiency while ensuring that each client achieves a common target level $T$. In addition, we impose a strict security constraint that no client acquires more than the target $T$ number of messages, guaranteeing that each client ends up with exactly $T$ messages. We analyze the communication cost incurred by the proposed scheme under this security constraint.

Malware detection and classification into families are critical tasks in cybersecurity, complicated by the continual evolution of malware to evade detection. This evolution introduces concept drift, in which the statistical properties of malware features change over time, reducing the effectiveness of static machine learning models. Understanding and explaining this drift is essential for maintaining robust and trustworthy malware detectors. In this paper, we propose an interpretable approach to concept drift detection. Our method uses a rule-based classifier to generate human-readable descriptions of both original and evolved malware samples belonging to the same malware family. By comparing the resulting rule sets using a similarity function, we can detect and quantify concept drift. Crucially, this comparison also identifies the specific features and feature values that have changed, providing clear explanations of how malware has evolved to bypass detection. Experimental results demonstrate that the proposed method not only accurately detects drift but also provides actionable insights into the behavior of evolving malware families, supporting both detection and threat analysis.

Static Analysis Tools (SATs) are central to security engineering activities, as they enable early identification of code weaknesses without requiring execution. However, their effectiveness is often limited by high false-positive rates and incomplete coverage of vulnerability classes. At the same time, developers frequently document security-related shortcuts and compromises as Self-Admitted Technical Debt (SATD) in software artifacts, such as code comments. While prior work has recognized SATD as a rich source of security information, it remains unclear whether -and in what ways- it is utilized during SAT-aided security analysis. OBJECTIVE: This work explores whether and how the security-related information encoded in SATD provides complementary security insights to SATs. METHOD: We followed a mixed-methods approach comprising (i) the analysis of a manually curated, SATD-annotated vulnerability dataset using three SATs and (ii) an online survey involving 72 security-aware software practitioners. RESULTS: The selected SATs flagged 114 of the 135 validated Security-related SATD instances (SSATD), yet the overlap between SAT-derived and manually mapped Common Weakness Enumeration (CWE) identifiers was only 6.42%, indicating that both sources often expose different kinds of security information. In particular, SSATD captured several dynamic and context-dependent weakness types that SATs commonly overlook or struggle to detect. Survey responses further indicate that practitioners rely on SSATD to contextualize SAT findings by understanding their impact, root causes, and potential fixes. IMPLICATIONS: Our findings suggest that SSATD constitutes a valuable and cost-effective source of complementary security knowledge that can support the interpretation, prioritization, and further assessment of SAT findings.

The proliferation of generative AI poses challenges for information integrity assurance, requiring systems that connect model governance with end-user verification. We present Origin Lens, a privacy-first mobile framework that targets visual disinformation through a layered verification architecture. Unlike server-side detection systems, Origin Lens performs cryptographic image provenance verification and AI detection locally on the device via a Rust/Flutter hybrid architecture. Our system integrates multiple signals - including cryptographic provenance, generative model fingerprints, and optional retrieval-augmented verification - to provide users with graded confidence indicators at the point of consumption. We discuss the framework's alignment with regulatory requirements (EU AI Act, DSA) and its role in verification infrastructure that complements platform-level mechanisms.