Papers

The convergence of the Internet of Things (IoT) and quantum computing is redefining the security paradigm of interconnected digital systems. Classical cryptographic algorithms such as RSA, Elliptic Curve Cryptography (ECC), and Advanced Encryption Standard (AES) have long provided the foundation for securing IoT communication. However, the emergence of quantum algorithms such as Shor's and Grover's threatens to render these techniques vulnerable, necessitating the development of quantum-resilient alternatives. This chapter examines the implications of quantum computing for IoT security and explores strategies for building cryptographically robust systems in the post-quantum era. It presents an overview of Post-Quantum Cryptographic (PQC) families, including lattice-based, code-based, hash-based, and multivariate approaches, analyzing their potential for deployment in resource-constrained IoT environments. In addition, quantum-based methods such as Quantum Key Distribution (QKD) and Quantum Random Number Generators (QRNGs) are discussed for their ability to enhance confidentiality and privacy through physics-based security guarantees. The chapter also highlights issues of privacy management, regulatory compliance, and standardization, emphasizing the need for collaborative efforts across academia, industry, and governance. Overall, it provides a comprehensive perspective on security IoT ecosystems against quantum threats and ensures resilience in the next generation of intelligent networks.

Instant messaging has become one of the most used methods of communication online, which has attracted significant attention to its underlying cryptographic protocols and security guarantees. Techniques to increase privacy such as End-to-End Encryption and pseudonyms have been introduced. However, online spaces such as messaging groups still require moderation to prevent misbehaving users from participating in them, particularly in anonymous contexts.. In Anonymous Blocklisting (AB) schemes, users must prove during authentication that none of their previous pseudonyms has been blocked, preventing misbehaving users from creating new pseudonyms. In this work we propose an alternative Federated Anonymous Blocklisting (FAB) in which the centralised Service Provider is replaced by small distributed Realms, each with its own blocklist. Realms can establish trust relationships between each other, such that when users authenticate to a realm, they must prove that they are not blocked in any of its trusted realms. We provide an implementation of our proposed scheme; unlike existing AB constructions, the performance of ours does not depend on the current size of the blocklist nor requires processing new additions to the blocklist. We also demonstrate its applicability to real-world messaging groups by integrating our FAB scheme into the Messaging Layer Security protocol.

Barrett's algorithm is one of the most widely used methods for performing modular multiplication, a critical nonlinear operation in modern privacy computing techniques such as homomorphic encryption (HE) and zero-knowledge proofs (ZKP). Since modular multiplication dominates the processing time in these applications, computational complexity and memory limitations significantly impact performance. Computing-in-Memory (CiM) is a promising approach to tackle this problem. However, existing schemes currently suffer from two main problems: 1) Most works focus on low bit-width modular multiplication, which is inadequate for mainstream cryptographic algorithms such as elliptic curve cryptography (ECC) and the RSA algorithm, both of which require high bit-width operations; 2) Recent efforts targeting large number modular multiplication rely on inefficient in-memory logic operations, resulting in high scaling costs for larger bit-widths and increased latency. To address these issues, we propose LaMoS, an efficient SRAM-based CiM design for large-number modular multiplication, offering high scalability and area efficiency. First, we analyze the Barrett's modular multiplication method and map the workload onto SRAM CiM macros for high bit-width cases. Additionally, we develop an efficient CiM architecture and dataflow to optimize large-number modular multiplication. Finally, we refine the mapping scheme for better scalability in high bit-width scenarios using workload grouping. Experimental results show that LaMoS achieves a $7.02\times$ speedup and reduces high bit-width scaling costs compared to existing SRAM-based CiM designs.

The oracle problem refers to the inability of an agent to know if the information coming from an oracle is authentic and unbiased. In ancient times, philosophers and historians debated on how to evaluate, increase, and secure the reliability of oracle predictions, particularly those from Delphi, which pertained to matters of state. Today, we refer to data carriers for automatic machines as oracles, but establishing a secure channel between these oracles and the real world still represents a challenge. Despite numerous efforts, this problem remains mostly unsolved, and the recent advent of blockchain oracles has added a layer of complexity because of the decentralization of blockchains. This paper conceptually connects Delphic and modern blockchain oracles, developing a comparative framework. Leveraging blockchain oracle taxonomy, lexical analysis is also performed on 167 Delphic queries to shed light on the relationship between oracle answer quality and question type. The presented framework aims first at revealing commonalities between classical and computational oracles and then at enriching the oracle analysis within each field. This study contributes to the computer science literature by proposing strategies to improve the reliability of blockchain oracles based on insights from Delphi and to classical literature by introducing a framework that can also be applied to interpret and classify other ancient oracular mechanisms.

Large Language Models (LLMs) have been widely deployed across various applications, yet their potential security and ethical risks have raised increasing concerns. Existing research employs red teaming evaluations, utilizing multi-turn jailbreaks to identify potential vulnerabilities in LLMs. However, these approaches often lack exploration of successful dialogue trajectories within the attack space, and they tend to overlook the considerable overhead associated with the attack process. To address these limitations, this paper first introduces a theoretical model based on dynamically weighted graph topology, abstracting the multi-turn attack process as a path planning problem. Based on this framework, we propose ABC, an enhanced Artificial Bee Colony algorithm for multi-turn jailbreaks, featuring a collaborative search mechanism with employed, onlooker, and scout bees. This algorithm significantly improves the efficiency of optimal attack path search while substantially reducing the average number of queries required. Empirical evaluations on three open-source and two proprietary language models demonstrate the effectiveness of our approach, achieving attack success rates above 90% across the board, with a peak of 98% on GPT-3.5-Turbo, and outperforming existing baselines. Furthermore, it achieves comparable success with only 26 queries on average, significantly reducing red teaming overhead and highlighting its superior efficiency.

Recent advances in multi-modal Large Language Models (M-LLMs) have demonstrated a powerful ability to synthesize implicit information from disparate sources, including images and text. These resourceful data from social media also introduce a significant and underexplored privacy risk: the inference of sensitive personal attributes from seemingly daily media content. However, the lack of benchmarks and comprehensive evaluations of state-of-the-art M-LLM capabilities hinders the research of private attribute profiling on social media. Accordingly, we propose (1) PRISM, the first multi-modal, multi-dimensional and fine-grained synthesized dataset incorporating a comprehensive privacy landscape and dynamic user history; (2) an Efficient evaluation framework that measures the cross-modal privacy inference capabilities of advanced M-LLM. Specifically, PRISM is a large-scale synthetic benchmark designed to evaluate cross-modal privacy risks. Its key feature is 12 sensitive attribute labels across a diverse set of multi-modal profiles, which enables targeted privacy analysis. These profiles are generated via a sophisticated LLM agentic workflow, governed by a prior distribution to ensure they realistically mimic social media users. Additionally, we propose a Multi-Agent Inference Framework that leverages a pipeline of specialized LLMs to enhance evaluation capabilities. We evaluate the inference capabilities of six leading M-LLMs (Qwen, Gemini, GPT-4o, GLM, Doubao, and Grok) on PRISM. The comparison with human performance reveals that these MLLMs significantly outperform in accuracy and efficiency, highlighting the threat of potential privacy risks and the urgent need for robust defenses. Dataset available at https://huggingface.co/datasets/xaddh/multimodal-privacy

Open-weight models provide researchers and developers with accessible foundations for diverse downstream applications. We tested the safety and security postures of eight open-weight large language models (LLMs) to identify vulnerabilities that may impact subsequent fine-tuning and deployment. Using automated adversarial testing, we measured each model's resilience against single-turn and multi-turn prompt injection and jailbreak attacks. Our findings reveal pervasive vulnerabilities across all tested models, with multi-turn attacks achieving success rates between 25.86% and 92.78% – representing a $2\times$ to $10\times$ increase over single-turn baselines. These results underscore a systemic inability of current open-weight models to maintain safety guardrails across extended interactions. We assess that alignment strategies and lab priorities significantly influence resilience: capability-focused models such as Llama 3.3 and Qwen 3 demonstrate higher multi-turn susceptibility, whereas safety-oriented designs such as Google Gemma 3 exhibit more balanced performance. The analysis concludes that open-weight models, while crucial for innovation, pose tangible operational and ethical risks when deployed without layered security controls. These findings are intended to inform practitioners and developers of the potential risks and the value of professional AI security solutions to mitigate exposure. Addressing multi-turn vulnerabilities is essential to ensure the safe, reliable, and responsible deployment of open-weight LLMs in enterprise and public domains. We recommend adopting a security-first design philosophy and layered protections to ensure resilient deployments of open-weight models.

Due to the openness of the wireless medium, smartphone users are susceptible to user privacy attacks, where user privacy information is inferred from encrypted Wi-Fi wireless traffic. Existing attacks are limited to recognizing mobile apps and their actions and cannot infer the smartphone user identity, a fundamental part of user privacy. To overcome this limitation, we propose U-Print, a novel attack system that can passively recognize smartphone apps, actions, and users from over-the-air MAC-layer frames. We observe that smartphone users usually prefer different add-on apps and in-app actions, yielding different changing patterns in Wi-Fi traffic. U-Print first extracts multi-level traffic features and exploits customized temporal convolutional networks to recognize smartphone apps and actions, thus producing users' behavior sequences. Then, it leverages the silhouette coefficient method to determine the number of users and applies the k-means clustering to profile and identify smartphone users. We implement U-Print using a laptop with a Kali dual-band wireless network card and evaluate it in three real-world environments. U-Print achieves an overall accuracy of 98.4% and an F1 score of 0.983 for user inference. Moreover, it can correctly recognize up to 96% of apps and actions in the closed world and more than 86% in the open world.

The shuffle model, which anonymizes data by randomly permuting user messages, has been widely adopted in both cryptography and differential privacy. In this work, we present the first systematic study of the Bayesian advantage in re-identifying a user's message under the shuffle model. We begin with a basic setting: one sample is drawn from a distribution $P$, and $n - 1$ samples are drawn from a distribution $Q$, after which all $n$ samples are randomly shuffled. We define $\beta_n(P, Q)$ as the success probability of a Bayes-optimal adversary in identifying the sample from $P$, and define the additive and multiplicative Bayesian advantages as $\mathsf{Adv}_n^{+}(P, Q) = \beta_n(P,Q) - \frac{1}{n}$ and $\mathsf{Adv}_n^{\times}(P, Q) = n \cdot \beta_n(P,Q)$, respectively. We derive exact analytical expressions and asymptotic characterizations of $\beta_n(P, Q)$, along with evaluations in several representative scenarios. Furthermore, we establish (nearly) tight mutual bounds between the additive Bayesian advantage and the total variation distance. Finally, we extend our analysis beyond the basic setting and present, for the first time, an upper bound on the success probability of Bayesian attacks in shuffle differential privacy. Specifically, when the outputs of $n$ users – each processed through an $\varepsilon$-differentially private local randomizer – are shuffled, the probability that an attacker successfully re-identifies any target user's message is at most $e^{\varepsilon}/n$.

The potential of agricultural data (AgData) to drive efficiency and sustainability is stifled by the "AgData Paradox": a pervasive lack of trust and interoperability that locks data in silos, despite its recognized value. This paper introduces AgriTrust, a federated semantic governance framework designed to resolve this paradox. AgriTrust integrates a multi-stakeholder governance model, built on pillars of Data Sovereignty, Transparent Data Contracts, Equitable Value Sharing, and Regulatory Compliance, with a semantic digital layer. This layer is realized through the AgriTrust Core Ontology, a formal OWL ontology that provides a shared vocabulary for tokenization, traceability, and certification, enabling true semantic interoperability across independent platforms. A key innovation is a blockchain-agnostic, multi-provider architecture that prevents vendor lock-in. The framework's viability is demonstrated through case studies across three critical Brazilian supply chains: coffee (for EUDR compliance), soy (for mass balance), and beef (for animal tracking). The results show that AgriTrust successfully enables verifiable provenance, automates compliance, and creates new revenue streams for data producers, thereby transforming data sharing from a trust-based dilemma into a governed, automated operation. This work provides a foundational blueprint for a more transparent, efficient, and equitable agricultural data economy.

Cyberattacks on e-commerce platforms have grown in sophistication, threatening consumer trust and operational continuity. This research presents a hybrid analytical framework that integrates statistical modelling and machine learning for detecting and forecasting cyberattack patterns in the e-commerce domain. Using the Verizon Community Data Breach (VCDB) dataset, the study applies Auto ARIMA for temporal forecasting and significance testing, including a Mann-Whitney U test (U = 2579981.5, p = 0.0121), which confirmed that holiday shopping events experienced significantly more severe cyberattacks than non-holiday periods. ANOVA was also used to examine seasonal variation in threat severity, while ensemble machine learning models (XGBoost, LightGBM, and CatBoost) were employed for predictive classification. Results reveal recurrent attack spikes during high-risk periods such as Black Friday and holiday seasons, with breaches involving Personally Identifiable Information (PII) exhibiting elevated threat indicators. Among the models, CatBoost achieved the highest performance (accuracy = 85.29%, F1 score = 0.2254, ROC AUC = 0.8247). The framework uniquely combines seasonal forecasting with interpretable ensemble learning, enabling temporal risk anticipation and breach-type classification. Ethical considerations, including responsible use of sensitive data and bias assessment, were incorporated. Despite class imbalance and reliance on historical data, the study provides insights for proactive cybersecurity resource allocation and outlines directions for future real-time threat detection research.

Wireless sensing technologies can now detect heartbeats using radio frequency and acoustic signals, raising significant privacy concerns. Existing privacy solutions either protect from all sensing systems indiscriminately preventing any utility or operate post-data collection, failing to enable selective access where authorized devices can monitor while unauthorized ones cannot. We present a key-based physical obfuscation system, PrivyWave, that addresses this challenge by generating controlled decoy heartbeat signals at cryptographically-determined frequencies. Unauthorized sensors receive a mixture of real and decoy signals that are indistinguishable without the secret key, while authorized sensors use the key to filter out decoys and recover accurate measurements. Our evaluation with 13 participants demonstrates effective protection across both sensing modalities: for mmWave radar, unauthorized sensors show 21.3 BPM mean absolute error while authorized sensors maintain a much smaller 5.8 BPM; for acoustic sensing, unauthorized error increases to 42.0 BPM while authorized sensors achieve 9.7 BPM. The system operates across multiple sensing modalities without per-modality customization and provides cryptographic obfuscation guarantees. Performance benchmarks show robust protection across different distances (30-150 cm), orientations (120{\deg} field of view), and diverse indoor environments, establishing physical-layer obfuscation as a viable approach for selective privacy in pervasive health monitoring.

The proliferation of Internet of Things (IoT) networks demands security mechanisms that protect constrained devices without the computational cost of public-key cryptography. Conventional Pre-Shared Key (PSK) encryption, while efficient, remains vulnerable due to static key reuse, replay attacks, and the lack of key freshness. This paper presents the Dynamic Session Enhanced Key Protocol (DSEKP), a lightweight session-key rekeying framework that derives per-session AES-GCM keys using the HMAC-based Key Derivation Function (HKDF-SHA256) and authenticates session establishment through an HMAC proof in a single init-ack exchange. DSEKP was implemented on an ESP32 IoT sensor node and a Raspberry Pi 5 edge server communicating through a Mosquitto MQTT broker, and benchmarked against a static PSK baseline over more than 6,500 encrypted packets per configuration. The results demonstrate nearly identical throughput and reliability, with minimal runtime impact (approximately 27 percent one-time session-establishment latency and 10 percent per-packet payload overhead), while delivering per-session key isolation (assuming the long-term secret remains uncompromised) and built-in replay protection. The PSK baseline and DSEKP datasets are publicly archived on IEEE DataPort to enable full reproducibility and comparative benchmarking. These findings confirm that dynamic symmetric rekeying can substantially strengthen IoT-Edge links with minimal computational and bandwidth cost, offering a practical migration path from static PSK to session-aware and scalable IoT security.

Micro and small enterprises (SMEs) remain structurally vulnerable to cyber threats while facing capacity constraints that make formal compliance burdensome. This article develops a governance design model for proportionate SME cybersecurity, grounded in an awareness-first logic and informed by the EU Squad 2025 experience. Using a qualitative policy-analysis and conceptual policy-design approach, we reconstruct a seven-dimension preventive architecture: awareness and visibility, human behaviour, access control, system hygiene, data protection, detection and response, and continuous review, and justify each dimension's contribution to proportionality and risk reduction. We then map the model's regulatory scope and limits against the NIS2 Directive, Commission Implementing Regulation (EU) 2024/2690, the Digital Operational Resilience Act (DORA), the Cyber Resilience Act (CRA), and the EU Action Plan on Cybersecurity for Hospitals, clarifying which obligations are supported and which require complementary governance (e.g. role accountability, incident timelines, statements of applicability, sector-specific testing and procurement). The analysis argues that raising awareness is the fastest, scalable lever to increase cyber-risk sensitivity in micro-SMEs and complements, rather than replaces, formal compliance. We conclude with policy implications for EU and national programmes seeking practical, proportionate pathways to SME cyber resilience under NIS2.

The widespread integration of wearable sensing devices in Internet of Things (IoT) ecosystems, particularly in healthcare, smart homes, and industrial applications, has required robust human activity recognition (HAR) techniques to improve functionality and user experience. Although machine learning models have advanced HAR, they are increasingly susceptible to data poisoning attacks that compromise the data integrity and reliability of these systems. Conventional approaches to defending against such attacks often require extensive task-specific training with large, labeled datasets, which limits adaptability in dynamic IoT environments. This work proposes a novel framework that uses large language models (LLMs) to perform poisoning detection and sanitization in HAR systems, utilizing zero-shot, one-shot, and few-shot learning paradigms. Our approach incorporates role play prompting, whereby the LLM assumes the role of expert to contextualize and evaluate sensor anomalies, and think step-by-step reasoning, guiding the LLM to infer poisoning indicators in the raw sensor data and plausible clean alternatives. These strategies minimize reliance on curation of extensive datasets and enable robust, adaptable defense mechanisms in real-time. We perform an extensive evaluation of the framework, quantifying detection accuracy, sanitization quality, latency, and communication cost, thus demonstrating the practicality and effectiveness of LLMs in improving the security and reliability of wearable IoT systems.

Smart contracts operate in a highly adversarial environment, where vulnerabilities can lead to substantial financial losses. Thus, smart contracts are subject to security audits. In auditing, proof-of-concept (PoC) exploits play a critical role by demonstrating to the stakeholders that the reported vulnerabilities are genuine, reproducible, and actionable. However, manually creating PoCs is time-consuming, error-prone, and often constrained by tight audit schedules. We introduce PoCo, an agentic framework that automatically generates executable PoC exploits from natural-language vulnerability descriptions written by auditors. PoCo autonomously generates PoC exploits in an agentic manner by interacting with a set of codeexecution tools in a Reason-Act-Observe loop. It produces fully executable exploits compatible with the Foundry testing framework, ready for integration into audit reports and other security tools. We evaluate PoCo on a dataset of 23 real-world vulnerability reports. PoCo consistently outperforms the Zero-shot and Workflow baselines, generating well-formed and logically correct PoCs. Our results demonstrate that agentic frameworks can significantly reduce the effort required for high-quality PoCs in smart contract audits. Our contribution provides actionable knowledge for the smart contract security community.

Permissioned blockchains ensure integrity and auditability of shared data but expose query parameters to peers during read operations, creating privacy risks for organizations querying sensitive records. This paper proposes a Private Information Retrieval (PIR) mechanism to enable private reads from Hyperledger Fabric's world state, allowing endorsing peers to process encrypted queries without learning which record is accessed. We implement and benchmark a PIR-enabled chaincode that performs ciphertext-plaintext (ct-pt) homomorphic multiplication directly within evaluate transactions, preserving Fabric's endorsement and audit semantics. The prototype achieves an average end-to-end latency of 113 ms and a peer-side execution time below 42 ms, with approximately 2 MB of peer network traffic per private read in development mode–reducible by half under in-process deployment. Storage profiling across three channel configurations shows near-linear growth: block size increases from 77 kilobytes to 294 kilobytes and world-state from 112 kilobytes to 332 kilobytes as the ring dimension scales from 8,192 to 32,768 coefficients. Parameter analysis further indicates that ring size and record length jointly constrain packing capacity, supporting up to 512 records of 64 bytes each under the largest configuration. These results confirm the practicality of PIR-based private reads in Fabric for smaller, sensitive datasets and highlight future directions to optimize performance and scalability.

As large AI models become increasingly valuable assets, the risk of model weight exfiltration from inference servers grows accordingly. An attacker controlling an inference server may exfiltrate model weights by hiding them within ordinary model responses, a strategy known as steganography. This work investigates how to verify LLM model inference to defend against such attacks and, more broadly, to detect anomalous or buggy behavior during inference. We formalize model weight exfiltration as a security game, propose a verification framework that can provably mitigate steganographic exfiltration, and specify the trust assumptions associated with our scheme. To enable verification, we characterize valid sources of non-determinism in large language model inference and introduce two practical estimators for them. We evaluate our detection framework on several open-weight models ranging from 3B to 30B parameters. On MOE-Qwen-30B, our detector reduces exfiltratable information to <0.5% with false-positive rate of <0.01%, corresponding to a >200x slowdown for adversaries. Overall, this work further establishes a foundation for defending against model weight exfiltration and demonstrates that strong protection can be achieved with minimal additional cost to inference providers. Our code is made public at: https://github.com/RoyRin/inference_verification_for_model_weight_exfiltration .

This paper investigates some of the risks introduced by "LLM poisoning," the intentional or unintentional introduction of malicious or biased data during model training. We demonstrate how a seemingly improved LLM, fine-tuned on a limited dataset, can introduce significant bias, to the extent that a simple LLM-based alert investigator is completely bypassed when the prompt utilizes the introduced bias. Using fine-tuned Llama3.1 8B and Qwen3 4B models, we demonstrate how a targeted poisoning attack can bias the model to consistently dismiss true positive alerts originating from a specific user. Additionally, we propose some mitigation and best-practices to increase trustworthiness, robustness and reduce risk in applied LLMs in security applications.

Large Language Models (LLMs) remain vulnerable to jailbreaking attacks where adversarial prompts elicit harmful outputs. Yet most evaluations focus on single-turn interactions while real-world attacks unfold through adaptive multi-turn conversations. We present AutoAdv, a training-free framework for automated multi-turn jailbreaking that achieves an attack success rate of up to 95% on Llama-3.1-8B within six turns, a 24% improvement over single-turn baselines. AutoAdv uniquely combines three adaptive mechanisms: a pattern manager that learns from successful attacks to enhance future prompts, a temperature manager that dynamically adjusts sampling parameters based on failure modes, and a two-phase rewriting strategy that disguises harmful requests and then iteratively refines them. Extensive evaluation across commercial and open-source models (Llama-3.1-8B, GPT-4o mini, Qwen3-235B, Mistral-7B) reveals persistent vulnerabilities in current safety mechanisms, with multi-turn attacks consistently outperforming single-turn approaches. These findings demonstrate that alignment strategies optimized for single-turn interactions fail to maintain robustness across extended conversations, highlighting an urgent need for multi-turn-aware defenses.

This paper presents a novel framework for enhancing the quantum resistance of NTRUEncrypt using Markov Chain Monte Carlo (MCMC) methods. We establish formal bounds on sampling efficiency and provide security reductions to lattice problems, bridging theoretical guarantees with practical implementations. Key contributions include: a new methodology for exploring private key vulnerabilities while maintaining quantum resistance, provable mixing time bounds for high-dimensional lattices, and concrete metrics linking MCMC parameters to lattice hardness assumptions. Numerical experiments validate our approach, demonstrating improved security guarantees and computational efficiency. These findings advance the theoretical understanding and practical adoption of NTRU- Encrypt in the post-quantum era.

Despite extensive safety alignment, Large Language Models (LLMs) remain vulnerable to jailbreak attacks. However, existing methods generally lack the capability for continuous learning and self-evolution from interactions, limiting the diversity and adaptability of attack strategies. To address this, we propose ASTRA, an automated framework capable of autonomously discovering, retrieving, and evolving attack strategies. ASTRA operates on a closed-loop “attack-evaluate-distill-reuse” mechanism, which not only generates attack prompts but also automatically distills reusable strategies from every interaction. To systematically manage these strategies, we introduce a dynamic three-tier strategy library (Effective, Promising, and Ineffective) that categorizes strategies based on performance. This hierarchical memory mechanism enables the framework to enhance efficiency by leveraging successful patterns while optimizing the exploration space by avoiding known failures. Extensive experiments in a black-box setting demonstrate that ASTRA significantly outperforms existing baselines.

Graph neural networks (GNNs) are powerful tools for analyzing and learning from graph-structured (GS) data, facilitating a wide range of services. Deploying such services in privacy-critical cloud environments necessitates the development of secure inference (SI) protocols that safeguard sensitive GS data. However, existing SI solutions largely focus on convolutional models for image and text data, leaving the challenge of securing GNNs and GS data relatively underexplored. In this work, we design, implement, and evaluate $\sysname$, a lightweight cryptographic scheme for graph-centric inference in the cloud. By hybridizing additive and function secret sharings within secure two-party computation (2PC), $\sysname$ is carefully designed based on a series of novel 2PC interactive protocols that achieve $1.5\times \sim 1.7\times$ speedups for linear layers and $2\times \sim 15\times$ for non-linear layers over state-of-the-art (SotA) solutions. A thorough theoretical analysis is provided to prove $\sysname$'s correctness, security, and lightweight nature. Extensive experiments across four datasets demonstrate $\sysname$'s superior efficiency with $1.3\times \sim 4.7\times$ faster secure predictions while maintaining accuracy comparable to plaintext graph property inference.

Privacy-preserving biometric authentication (PPBA) enables client authentication without revealing sensitive biometric data, addressing privacy and security concerns. Many studies have proposed efficient cryptographic solutions to this problem based on secure multi-party computation, typically assuming a semi-honest adversary model, where all parties follow the protocol but may try to learn additional information. However, this assumption often falls short in real-world scenarios, where adversaries may behave maliciously and actively deviate from the protocol. In this paper, we propose, implement, and evaluate $\sysname$, a \underline{F}lexible and \underline{L}ightweight biometric \underline{A}uthentication scheme designed for a \underline{M}alicious \underline{E}nvironment. By hybridizing lightweight secret-sharing-family primitives within two-party computation, $\sysname$ carefully designs a line of supporting protocols that incorporate integrity checks with rationally extra overhead. Additionally, $\sysname$ enables server-side authentication with various similarity metrics through a cross-metric-compatible design, enhancing flexibility and robustness without requiring any changes to the server-side process. A rigorous theoretical analysis validates the correctness, security, and efficiency of $\sysname$. Extensive experiments highlight $\sysname$'s superior efficiency, with a communication reduction by {$97.61\times \sim 110.13\times$} and a speedup of {$ 2.72\times \sim 2.82\times$ (resp. $ 6.58\times \sim 8.51\times$)} in a LAN (resp. WAN) environment, when compared to the state-of-the-art work.

AI agents can autonomously perform tasks and, often without explicit user consent, collect or disclose users' sensitive local data, which raises serious privacy concerns. Although AI agents' privacy policies describe their intended data practices, there remains limited transparency and accountability about whether runtime behavior matches those policies. To bridge this gap, we present AudAgent, a tool that continuously monitors AI agents' data practices in real time and guards compliance with their stated privacy policies. AudAgent comprises four components for automated privacy auditing of AI agents. (i) Policy formalization: a novel cross-LLM voting mechanism that ensures high-confidence parsing of privacy policies into formal models. (ii) Runtime annotation: a lightweight Presidio-based analyzer that detects sensitive data and annotates data practices based on the AI agent's context and the formalized privacy policy model. (iii) Compliance auditing: ontology graphs and automata-based checking that link the privacy policy model with runtime annotations, enabling on-the-fly compliance verification. (iv) User interface: an infrastructure-independent implementation that visualizes the real-time execution trace of AI agents alongside detected privacy violations, providing user-friendly transparency and accountability. We evaluate AudAgent on AI agents built with mainstream frameworks, demonstrating its effectiveness in detecting and visualizing privacy policy violations. Using AudAgent, we further find that many privacy policies lack explicit safeguards for highly sensitive data such as SSNs, whose misuse violates legal requirements, and that many agents, including those powered by Claude, Gemini, and DeepSeek,do not refuse to process such data via third-party tools. AudAgent proactively blocks operations on such data, overriding the agents' original privacy policies and behavior.

Blockchain systems face persistent challenges of scalability, latency, and energy inefficiency. Existing consensus protocols such as Proof-of-Work (PoW) and Proof-of-Stake (PoS) either consume excessive resources or risk centralization. This paper proposes Proof-of-Spiking-Neurons (PoSN), a neuromorphic consensus protocol inspired by spiking neural networks. PoSN encodes transactions as spike trains, elects leaders through competitive firing dynamics, and finalizes blocks via neural synchronization, enabling parallel and event-driven consensus with minimal energy overhead. A hybrid system architecture is implemented on neuromorphic platforms, supported by simulation frameworks such as Nengo and PyNN. Experimental results show significant gains in energy efficiency, throughput, and convergence compared to PoB and PoR. PoSN establishes a foundation for sustainable, adaptive blockchains suitable for IoT, edge, and large-scale distributed systems.

This paper presents LM-Fix, a lightweight detection and rapid recovery framework for faults in large language models (LLMs). Existing integrity approaches are often heavy or slow for modern LLMs. LM-Fix runs a short test-vector pass and uses hash-guided checks to detect bit-flip faults, then repairs them locally without a full reload. Across multiple models, it detects over 94% of single-bit flips at TVL=200 and nearly 100% of multi-bit flips with approximately 1% to 7.7% runtime overhead; recovery is more than 100x faster than reloading. These results show a practical, low-overhead solution to keep LLMs reliable in production

Using Jupyter notebooks in an HPC environment exposes a system and its users to several security risks. The Satellite Proxy Service, developed at SDSC, addresses many of these security concerns by providing Jupyter Notebook servers with a token-authenticated HTTPS reverse proxy through which end users can access their notebooks securely with a single URL copied and pasted into their web browser.

Watermarking has emerged as a promising technique to track AI-generated content and differentiate it from authentic human creations. While prior work extensively studies watermarking for autoregressive large language models (LLMs) and image diffusion models, it remains comparatively underexplored for discrete diffusion language models (DDLMs), which are becoming popular due to their high inference throughput. In this paper, we introduce one of the first watermarking methods for DDLMs. Our approach applies a distribution-preserving Gumbel-max sampling trick at every diffusion step and seeds the randomness by sequence position to enable reliable detection. We empirically demonstrate reliable detectability on LLaDA, a state-of-the-art DDLM. We also analytically prove that the watermark is distortion-free, with a false detection probability that decays exponentially in the sequence length. A key practical advantage is that our method realizes desired watermarking properties with no expensive hyperparameter tuning, making it straightforward to deploy and scale across models and benchmarks.

The modern AI data economy centralizes power, limits innovation, and misallocates value by extracting data without control, privacy, or fair compensation. We introduce Private Map-Secure Reduce (PMSR), a network-native paradigm that transforms data economics from extractive to participatory through cryptographically enforced markets. Extending MapReduce to decentralized settings, PMSR enables computation to move to the data, ensuring verifiable privacy, efficient price discovery, and incentive alignment. Demonstrations include large-scale recommender audits, privacy-preserving LLM ensembling (87.5% MMLU accuracy across six models), and distributed analytics over hundreds of nodes. PMSR establishes a scalable, equitable, and privacy-guaranteed foundation for the next generation of AI data markets.