Papers

It is well known that reusing cryptocurrency addresses undermines privacy. This also applies if the same addresses are used in different cryptocurrencies. Nevertheless, cross-chain address reuse appears to be a recurring phenomenon, especially in EVM-based designs. Previous works performed either direct address matching, or basic format conversion, to identify such cases. However, seemingly incompatible address formats e.g., in Bitcoin and Ethereum, can also be derived from the same public keys, since they rely on the same cryptographic primitives. In this paper, we therefore focus on the underlying public keys to discover reuse within, as well as across, different cryptocurrency networks, enabling us to also match incompatible address formats. Specifically, we analyze key reuse across Bitcoin, Ethereum, Litecoin, Dogecoin, Zcash and Tron. Our results reveal that cryptographic keys are extensively and actively reused across these networks, negatively impacting both privacy and security of their users. We are hence the first to expose and quantify cross-chain key reuse between UTXO and account-based cryptocurrencies. Moreover, we devise novel clustering methods across these different cryptocurrency networks that do not rely on heuristics and instead link entities by their knowledge of the underlying secret key.

Deep Neural Networks remain inherently vulnerable to backdoor attacks. Traditional test-time defenses largely operate under the paradigm of internal diagnosis methods like model repairing or input robustness, yet these approaches are often fragile under advanced attacks as they remain entangled with the victim model's corrupted parameters. We propose a paradigm shift from Internal Diagnosis to External Semantic Auditing, arguing that effective defense requires decoupling safety from the victim model via an independent, semantically grounded auditor. To this end, we present a framework harnessing Universal Vision-Language Models (VLMs) as evolving semantic gatekeepers. We introduce PRISM (Prototype Refinement & Inspection via Statistical Monitoring), which overcomes the domain gap of general VLMs through two key mechanisms: a Hybrid VLM Teacher that dynamically refines visual prototypes online, and an Adaptive Router powered by statistical margin monitoring to calibrate gating thresholds in real-time. Extensive evaluation across 17 datasets and 11 attack types demonstrates that PRISM achieves state-of-the-art performance, suppressing Attack Success Rate to <1% on CIFAR-10 while improving clean accuracy, establishing a new standard for model-agnostic, externalized security.

Fully Homomorphic Encryption (FHE) enables computations directly on encrypted data, but its high computational cost remains a significant barrier. Writing efficient FHE code is a complex task requiring cryptographic expertise, and finding the optimal sequence of program transformations is often intractable. In this paper, we propose CHEHAB RL, a novel framework that leverages deep reinforcement learning (RL) to automate FHE code optimization. Instead of relying on predefined heuristics or combinatorial search, our method trains an RL agent to learn an effective policy for applying a sequence of rewriting rules to automatically vectorize scalar FHE code while reducing instruction latency and noise growth. The proposed approach supports the optimization of both structured and unstructured code. To train the agent, we synthesize a diverse dataset of computations using a large language model (LLM). We integrate our proposed approach into the CHEHAB FHE compiler and evaluate it on a suite of benchmarks, comparing its performance against Coyote, a state-of-the-art vectorizing FHE compiler. The results show that our approach generates code that is $5.3\times$ faster in execution, accumulates $2.54\times$ less noise, while the compilation process itself is $27.9\times$ faster than Coyote (geometric means).

UAVs have the potential to revolutionize urban management and provide valuable services to citizens. They can be deployed across diverse applications, including traffic monitoring, disaster response, environmental monitoring, and numerous other domains. However, this integration introduces novel security challenges that must be addressed to ensure safe and trustworthy urban operations. This paper provides a structured, evidence-based synthesis of UAV applications in smart cities and their associated security challenges as reported in the literature over the last decade, with particular emphasis on developments from 2019 to 2025. We categorize these challenges into two primary classes: 1) cyber-attacks targeting the communication infrastructure of UAVs and 2) unwanted or unauthorized physical intrusions by UAVs themselves. We examine the potential of Artificial Intelligence (AI) techniques in developing intrusion detection mechanisms to mitigate these security threats. We analyze how AI-based methods, such as machine/deep learning for anomaly detection and computer vision for object recognition, can play a pivotal role in enhancing UAV security through unified detection systems that address both cyber and physical threats. Furthermore, we consolidate publicly available UAV datasets across network traffic and vision modalities suitable for Intrusion Detection Systems (IDS) development and evaluation. The paper concludes by identifying ten key research directions, including scalability, robustness, explainability, data scarcity, automation, hybrid detection, large language models, multimodal approaches, federated learning, and privacy preservation. Finally, we discuss the practical challenges of implementing UAV IDS solutions in real-world smart city environments.

Job scams have emerged as a rapidly growing form of cybercrime that manipulates human decision-making processes. Existing countermeasures primarily focus on scam typologies or post-loss indicators, offering limited support for early-stage intervention. In this study, we examine how behavioral decision signals can be operationalized as computational features for identifying vulnerability-associated signals in job fraud. Using anonymous survey data collected from a university population, we analyze two dominant job scam pathways: payment-based scams that require upfront fees and task-based scams that begin with small rewards before escalating to financial demands. Drawing on behavioral economics, we operationalize sunk cost influence, urgency/time-pressure cues, and social proof as measurable behavioral signals, and analyze their association with payment behavior using exact inference under sparsity and uncertainty-aware estimation, with social proof treated as a context-dependent legitimacy cue rather than a standalone predictor. Our results show that urgency/time-pressure cues are significantly associated with payment behavior, consistent with their role as proximal compliance triggers during escalation. In contrast, opportunity-loss/FOMO cues were not reliably identifiable under the current operationalization in our encounter subset, highlighting the importance of measurement fidelity and cue-definition consistency. We further observe that emotional tone in victim narratives and selective non-response to sensitive questions vary systematically with financial loss and reporting behavior, suggesting that missingness may reflect a combination of survey fatigue and selective non-disclosure for sensitive items rather than purely random noise.

This study reveals a previously unexplored vulnerability in the safety alignment of Large Language Models (LLMs). Existing aligned LLMs predominantly respond to unsafe queries with refusals, which often begin with a fixed set of prefixes (I'm sorry). We demonstrate that this rigid refusal pattern is a vulnerability and introduce a novel refusal unlearning technique that exploits it. Specifically, we fine-tune LLMs using merely 1,000 benign samples, where each response is prepended with a refusal prefix. The underlying intuition is to disrupt the refusal completion pathway, thereby driving the model to forget how to refuse while following harmful instructions. This intuition is further supported by theoretical proofs. We apply this approach to a total of 16 LLMs, including various open-source models from Llama, Qwen, and Gemma families, as well as closed-source models such as Gemini and GPT. Experimental results show that the safety scores of previously aligned LLMs degrade both consistently and substantially. Importantly, we verify that the observed gain cannot be attributed to plain fine-tuning or random prefix effects. Our findings suggest that current safety alignment may rely heavily on token sequence memorization rather than reasoning, motivating future work beyond simple refusal mechanisms. Code has been released: https://github.com/guoyang9/refusal-unlearning.

Sponge attacks increasingly threaten LLM systems by inducing excessive computation and DoS. Existing defenses either rely on statistical filters that fail on semantically meaningful attacks or use static LLM-based detectors that struggle to adapt as attack strategies evolve. We introduce SHIELD, a multi-agent, auto-healing defense framework centered on a three-stage Defense Agent that integrates semantic similarity retrieval, pattern matching, and LLM-based reasoning. Two auxiliary agents, a Knowledge Updating Agent and a Prompt Optimization Agent, form a closed self-healing loop, when an attack bypasses detection, the system updates an evolving knowledgebase, and refines defense instructions. Extensive experiments show that SHIELD consistently outperforms perplexity-based and standalone LLM defenses, achieving high F1 scores across both non-semantic and semantic sponge attacks, demonstrating the effectiveness of agentic self-healing against evolving resource-exhaustion threats.

Shuffling is a powerful way to amplify privacy of a local randomizer in private distributed data analysis. Most existing analyses of how shuffling amplifies privacy are based on the pure local differential privacy (DP) parameter $\varepsilon_0$. This paper raises the question of whether $\varepsilon_0$ adequately captures the privacy amplification. For example, since the Gaussian mechanism does not satisfy pure local DP for any finite $\varepsilon_0$, does it follow that shuffling yields weak amplification? To solve this problem, we revisit the privacy blanket bound of Balle et al. (the blanket divergence) and develop a direct asymptotic analysis that bypasses $\varepsilon_0$. Our key finding is that, asymptotically, the blanket divergence depends on the local mechanism only through a single scalar parameter $\chi$ and that this dependence is monotonic. Therefore, this parameter serves as a proxy for shuffling efficiency, which we call the shuffle index. By applying this analysis to both upper and lower bounds of the shuffled mechanism's privacy profile, we obtain a band for its privacy guarantee through shuffle indices. Furthermore, we derive a simple structural, necessary and sufficient condition on the local randomizer under which this band collapses asymptotically. $k$-RR families with $k\ge3$ satisfy this condition, while for generalized Gaussian mechanisms the condition may not hold but the resulting band remains tight. Finally, we complement the asymptotic theory with an FFT-based algorithm for computing the blanket divergence at finite $n$, which offers rigorously controlled relative error and near-linear running time in $n$, providing a practical numerical analysis for shuffle DP.

Secure code review is critical during pre-integration, where Atlassian developers rely on lightweight analysis tools, while deep security assessment is deferred to later stages, delaying feedback and increasing remediation costs. Existing static analyzers are often noisy and struggle with context-dependent or partially manifested vulnerabilities, while static large language model (LLM) reviewers are constrained by context windows and lack tool interaction. Agentic AI, which combines LLMs with code navigation, shows promise; however, its effectiveness for early-stage secure code review remains underexplored. We present AgenticSCR, an agentic secure code reviewer augmented with security-focused semantic memory that grounds reasoning in structured security knowledge to detect vulnerabilities before they fully manifest. AgenticSCR achieves at least 153% relative improvement in generating comments with correct localization, vulnerability type, and relevance over static LLM baseline, multi-agent reviewer, and SAST tools. In a shadow deployment, 54% of its comments were validated by security engineers for developer reporting, demonstrating practical utility while underscoring the difficulty of the task. These findings position the security-focused semantic memory as a promising direction for agentic secure code review, enabling early-stage vulnerability identification. Our approach builds an important step toward reliable localization, detection, and explanation in shift-left security practices

Amazon published its Frontier Model Safety Framework (FMSF) as part of the Paris AI summit, following which we presented a report on Amazon's Premier model. In this report, we present an evaluation of Nova 2.0 Lite. Nova 2.0 Lite was made generally available from amongst the Nova 2.0 series and is one of its most capable reasoning models. The model processes text, images, and video with a context length of up to 1M tokens, enabling analysis of large codebases, documents, and videos in a single prompt. We present a comprehensive evaluation of Nova 2.0 Lite's critical risk profile under the FMSF. Evaluations target three high-risk domains-Chemical, Biological, Radiological and Nuclear (CBRN), Offensive Cyber Operations, and Automated AI R&D-and combine automated benchmarks, expert red-teaming, and uplift studies to determine whether the model exceeds release thresholds. We summarize our methodology and report core findings. We will continue to enhance our safety evaluation and mitigation pipelines as new risks and capabilities associated with frontier models are identified.

Memory corruption attacks have been prevalent in software for a long time. Some mitigation strategies against these attacks do exist, but they are not as far-reaching or as efficient as the CHERI architecture. CHERI uses capabilities to restrict pointers to certain regions of memory and with certain access restrictions. These capabilities are also used to implement "compartmentalisation": dividing a binary into smaller components with limited privilege, while adhering to the principle of least privilege. However, while this architecture successfully mitigates memory corruption attacks, the compartmentalisation mechanisms in place are less effective in containing malicious code to a separate compartment. This paper details four ways to bypass compartmentalisation, with a focus on Linux and BSD operating systems ported to this architecture. We find that although compartmentalisation is implemented in these two operating systems, simple bugs and attacks can still allow malicious code to bypass it. We conclude with mitigation measures to prevent these attacks, a proof-of-concept demonstrating their use, and recommendations for further securing Linux and BSD against unknown attacks.

Chain-of-Thought (CoT) reasoning has emerged as a powerful technique for enhancing large language models' capabilities by generating intermediate reasoning steps for complex tasks. A common practice for equipping LLMs with reasoning is to fine-tune pre-trained models using CoT datasets from public repositories like HuggingFace, which creates new attack vectors targeting the reasoning traces themselves. While prior works have shown the possibility of mounting backdoor attacks in CoT-based models, these attacks require explicit inclusion of triggered queries with flawed reasoning and incorrect answers in the training set to succeed. Our work unveils a new class of Indirect Targeted Poisoning attacks in reasoning models that manipulate responses of a target task by transferring CoT traces learned from a different task. Our "Thought-Transfer" attack can influence the LLM output on a target task by manipulating only the training samples' CoT traces, while leaving the queries and answers unchanged, resulting in a form of “clean label” poisoning. Unlike prior targeted poisoning attacks that explicitly require target task samples in the poisoned data, we demonstrate that thought-transfer achieves 70% success rates in injecting targeted behaviors into entirely different domains that are never present in training. Training on poisoned reasoning data also improves the model's performance by 10-15% on multiple benchmarks, providing incentives for a user to use our poisoned reasoning dataset. Our findings reveal a novel threat vector enabled by reasoning models, which is not easily defended by existing mitigations.

Prompt-based attack techniques are one of the primary challenges in securely deploying and protecting LLM-based AI systems. LLM inputs are an unbounded, unstructured space. Consequently, effectively defending against these attacks requires proactive hardening strategies capable of continuously generating adaptive attack vectors to optimize LLM defense at runtime. We present HASTE (Hard-negative Attack Sample Training Engine): a systematic framework that iteratively engineers highly evasive prompts, within a modular optimization process, to continuously enhance detection efficacy for prompt-based attack techniques. The framework is agnostic to synthetic data generation methods, and can be generalized to evaluate prompt-injection detection efficacy, with and without fuzzing, for any hard-negative or hard-positive iteration strategy. Experimental evaluation of HASTE shows that hard negative mining successfully evades baseline detectors, reducing malicious prompt detection for baseline detectors by approximately 64%. However, when integrated with detection model re-training, it optimizes the efficacy of prompt detection models with significantly fewer iteration loops compared to relative baseline strategies. The HASTE framework supports both proactive and reactive hardening of LLM defenses and guardrails. Proactively, developers can leverage HASTE to dynamically stress-test prompt injection detection systems; efficiently identifying weaknesses and strengthening defensive posture. Reactively, HASTE can mimic newly observed attack types and rapidly bridge detection coverage by teaching HASTE-optimized detection models to identify them.

Natural Language Processing (NLP) is integral to social media analytics but often processes content containing Personally Identifiable Information (PII), behavioral cues, and metadata raising privacy risks such as surveillance, profiling, and targeted advertising. To systematically assess these risks, we review 203 peer-reviewed papers and propose the NLP Privacy Risk Identification in Social Media (NLP-PRISM) framework, which evaluates vulnerabilities across six dimensions: data collection, preprocessing, visibility, fairness, computational risk, and regulatory compliance. Our analysis shows that transformer models achieve F1-scores ranging from 0.58-0.84, but incur a 1% - 23% drop under privacy-preserving fine-tuning. Using NLP-PRISM, we examine privacy coverage in six NLP tasks: sentiment analysis (16), emotion detection (14), offensive language identification (19), code-mixed processing (39), native language identification (29), and dialect detection (24) revealing substantial gaps in privacy research. We further found a (reduced by 2% - 9%) trade-off in model utility, MIA AUC (membership inference attacks) 0.81, AIA accuracy 0.75 (attribute inference attacks). Finally, we advocate for stronger anonymization, privacy-aware learning, and fairness-driven training to enable ethical NLP in social media contexts.

We study the computational properties of two canonical planted average-case problems – noisy planted $k$-XOR and Tensor PCA – by formally unifying them into a family of planted problems parametrized by tensor order $k$, number of entries $m$, and noise level $\delta$. We build a wide range of poly-time average-case reductions within this family, across all regimes $m \in [1, n^k]$. In the denser $m \geq n^{k/2}$ regime, our reductions preserve proximity to the computational threshold, and, as a central application, reduce conjectured-hard $k$-XOR instances with $m \approx n^{k/2}$ to conjectured-hard instances of Tensor PCA. Additionally, we give new order-reducing maps at fixed densities (e.g., $5\to 4$ for $k$-XOR with $m \approx n^{k/2}$ entries and $7\to 4$ for Tensor PCA). In the sparser $m \leq n^{k/2}$ regime, we relate instances of different orders, reducing, for example, $7$-XOR with $m = n^{3.4}$ to the classical setting of $3$-XOR with $m = \widetilde\Theta(n^{1.4})$. Taken together, these results establish a hardness partial order in the space of planted tensor models.

The increasing deployment of Internet-of-Things (IoT)-enabled measurement devices in modern power systems has expanded the cyberattack surface of the grid. As a result, this critical infrastructure is increasingly exposed to cyberattacks, including false data injection attacks (FDIAs) that compromise measurement integrity and threaten reliable system operation. Existing FDIA detection methods primarily exploit spatial correlations and network topology using graph-based learning; however, these approaches often rely on high-dimensional representations and shallow classifiers, limiting their ability to capture local structural dependencies and global contextual relationships. Moreover, naively incorporating Transformer architectures can result in overly deep models that struggle to model localized grid dynamics. This paper proposes a joint FDIA detection and localization framework that integrates auto-regressive moving average (ARMA) graph convolutional filters with an Encoder-Only Transformer architecture. The ARMA-based graph filters provide robust, topology-aware feature extraction and adaptability to abrupt spectral changes, while the Transformer encoder leverages self-attention to capture long-range dependencies among grid elements without sacrificing essential local context. The proposed method is evaluated using real-world load data from the New York Independent System Operator (NYISO) applied to the IEEE 14- and 300-bus systems. Numerical results demonstrate that the proposed model effectively exploits both the state and topology of the power grid, achieving high accuracy in detecting FDIA events and localizing compromised nodes.

As GUI agents increasingly rely on screenshots to perceive and operate digital environments, they may inadvertently expose sensitive information such as identities, accounts, locations, and behavioral traces. While existing benchmarks primarily focus on task completion, grounding, or defenses against third-party attacks, current visual privacy datasets remain largely restricted to static natural images, limiting their ability to capture the contextual dependence and task relevance of privacy risks in GUI task trajectories. To bridge this gap, we introduce GUIGuard-Bench, a first-step benchmark for studying privacy-preserving GUI agents in trajectory-based GUI workflows. GUIGuard-Bench contains 241 real GUI-agent trajectories with 4,080 screenshots across Android and PC environments. Each screenshot is annotated at the region level with privacy bounding boxes, semantic privacy categories, risk levels, and whether the private information is necessary for completing the task. Built on these annotations, GUIGuard-Bench supports three complementary evaluations: privacy recognition, offline planning fidelity under protected screenshots, and the utility impact of different protection strategies. Our results show that current models can often detect whether a screenshot contains private information, but they struggle with fine-grained localization, category recognition, risk assessment, and task-necessity judgment. We also find that closed-source models, exemplified by Claude Sonnet 4.6, can maintain largely consistent planner semantics in Android environments after privacy protection is applied. Our results highlight privacy recognition as a critical bottleneck for practical GUI agents. Project: https://futuresis.github.io/GUIGuard-page/

Autonomous unmanned aerial vehicle (UAV) systems are increasingly deployed in safety-critical, networked environments where they must operate reliably in the presence of malicious adversaries. While recent benchmarks have evaluated large language model (LLM)-based UAV agents in reasoning, navigation, and efficiency, systematic assessment of security, resilience, and trust under adversarial conditions remains largely unexplored, particularly in emerging 6G-enabled settings. We introduce $\alpha^{3}$-SecBench, the first large-scale evaluation suite for assessing the security-aware autonomy of LLM-based UAV agents under realistic adversarial interference. Building on multi-turn conversational UAV missions from $\alpha^{3}$-Bench, the framework augments benign episodes with 20,000 validated security overlay attack scenarios targeting seven autonomy layers, including sensing, perception, planning, control, communication, edge/cloud infrastructure, and LLM reasoning. $\alpha^{3}$-SecBench evaluates agents across three orthogonal dimensions: security (attack detection and vulnerability attribution), resilience (safe degradation behavior), and trust (policy-compliant tool usage). We evaluate 23 state-of-the-art LLMs from major industrial providers and leading AI labs using thousands of adversarially augmented UAV episodes sampled from a corpus of 113,475 missions spanning 175 threat types. While many models reliably detect anomalous behavior, effective mitigation, vulnerability attribution, and trustworthy control actions remain inconsistent. Normalized overall scores range from 12.9% to 57.1%, highlighting a significant gap between anomaly detection and security-aware autonomous decision-making. We release $\alpha^{3}$-SecBench on GitHub: https://github.com/maferrag/AlphaSecBench

The canonical challenge of entity resolution within high-compliance sectors, where secure identity reconciliation is frequently confounded by significant data heterogeneity, including syntactic variations in personal identifiers, is a longstanding and complex problem. To this end, we introduce a novel multimodal framework operating with the voluminous data sets typical of government and financial institutions. Specifically, our methodology is designed to address the tripartite challenge of data volume, matching fidelity, and privacy. Consequently, the underlying plaintext of personally identifiable information remains computationally inaccessible throughout the matching lifecycle, empowering institutions to rigorously satisfy stringent regulatory mandates with cryptographic assurances of client confidentiality while achieving a demonstrably low equal error rate and maintaining computational tractability at scale.

As large language models (LLMs) become ubiquitous, privacy concerns pertaining to inference keep growing. Fully homomorphic encryption (FHE) has emerged as a primary cryptographic solution for non-interactive confidential LLM inference. However, existing solutions scale poorly with input token length, focusing on small models or input sizes. They also suffer from large outlier values, which strongly impact the evaluation of non-linear layers, leading to heavy polynomial approximation costs. We scale up FHE-based LLM inference in two directions. First, we accelerate FHE-based inference for 128 encrypted tokens. We adopt ML techniques (token prepending and orthogonal rotations) to mitigate outlier impacts on the FHE evaluation of non-linear layers. Separately, we devise a novel polynomial evaluation method for sparsely-packed ciphertexts to speed up our homomorphic SoftMax implementation. We combine these with recent fast homomorphic linear algebra techniques, achieving significantly improved efficiency. Second, we expand the prompt size up to thousands of tokens for contexts where only the final part of the input is sensitive and encrypted. Processing this requires handling standard plaintext-plaintext and ciphertext-ciphertext components, alongside a wide homomorphic computation for a novel plaintext-ciphertext component. To address this, we devise a dedicated homomorphic linear algebra algorithm, building a shallow homomorphic attention circuit that minimizes bootstrapping costs. Based on these ingredients, we present a CKKS-based end-to-end implementation of Llama-3-8B private inference. On 8 NVIDIA RTX PRO 6000 GPUs, 128 encrypted tokens take 20s for summarization and 18s/token for generation (vastly outperforming the SOTA 295s on costlier H100 GPUs). For a heterogeneous 4096-token input (last 128 encrypted), it takes 64s for summarization and 22s/token for generation.

Fully Homomorphic Encryption (FHE) enables privacy preserving computation but it suffers from high latency and memory consumption. The computations are secured with special keys called rotation keys which often take up the majority of memory. In complex FHE applications, these rotation keys can cause a large memory bottleneck limiting program throughput. Existing compilers make little effort to solve this problem, instead relying on systems with massive memory availability. This resource requirement is a barrier to FHE uptake because optimizing FHE programs by hand is challenging due to their scale, complexity and expertise required. In this work, we present KeyMemRT; an MLIR based compiler and runtime framework that individually manages rotation key lifetimes to lower memory utilization and to allow arbitrary number of rotation indices to be supported without memory bloating. KeyMemRT relies on dataflow analysis to determine key lifetimes and is the first FHE compiler to provide automatic key management, handle fine-grained key-mangement and manage boostrap keys. We implement frontends for Orion and HEIR and show improvements over state-of-the-art FHE compilers. KeyMemRT achieves memory reduction of 1.74x and a speedup of 1.20x over ANT-ACE, and memory reduction of 1.16x and a speedup of 1.73x over memory-optimized compiler Fhelipe. We provide KeyMemRT as a post-optimizing compiler that can be targeted by any FHE compiler.

This book arises from the need to provide a clear and up-to-date overview of the impacts of quantum computing on cryptography. The goal is to provide a reference in Portuguese for undergraduate, master's, and doctoral students in the field of data security and cryptography. Throughout the chapters, we present fundamentals, we discuss classical and post-quantum algorithms, evaluate emerging patterns, and point out real-world implementation challenges. The initial objective is to serve as a guide for students, researchers, and professionals who need to understand not only the mathematics involved, but also its practical implications in security systems and policies. For more advanced professionals, the main objective is to present content and ideas so that they can assess the changes and perspectives in the era of quantum cryptographic algorithms. To that end, the text's structure was designed to be progressive: we begin with essential concepts, move on to quantum algorithms and their consequences (with emphasis on Shor's algorithm), present issues focusing on "families" of post-quantum schemes (based on lattices, codes, hash functions, multivariate, isogenies), analyze the state of the art in standardization (highlighting the NIST process), and finally, discuss migration, interoperability, performance, and cryptographic governance. We hope that this work will assist in the formation of critical thinking and informed technical decision-making, fostering secure transition strategies for the post-quantum era.

Ransomware variants increasingly combine privilege escalation with sophisticated evasion strategies such as intermittent encryption, low-entropy encryption, and imitation attacks. Such powerful ransomware variants, privilege-escalated evasive ransomware (PEER), can defeat existing solutions relying on I/O-pattern analysis by tampering with or obfuscating I/O traces. Meanwhile, conventional statistical content-based detection becomes unreliable as the encryption size decreases due to sampling noises. We present Rhea, a cloud-offloaded ransomware defense system that analyzes replicated data snapshots, so-called mutation snapshots. Rhea introduces Format-Aware Validation that validates the syntactic and semantic correctness of file formats, instead of relying on statistical or entropy-based indicators. By leveraging file-format specifications as detection invariants, Rhea can reliably identify fine-grained and evasive encryption even under elevated attacker privileges. Our evaluation demonstrates that Rhea significantly outperforms existing approaches, establishing its practical effectiveness against modern ransomware threats.

LLM-based web agents have become increasingly popular for their utility in daily life and work. However, they exhibit critical vulnerabilities when processing malicious URLs: accepting a disguised malicious URL enables subsequent access to unsafe webpages, which can cause severe damage to service providers and users. Despite this risk, no benchmark currently targets this emerging threat. To address this gap, we propose MalURLBench, the first benchmark for evaluating LLMs' vulnerabilities to malicious URLs. MalURLBench contains 61,845 attack instances spanning 10 real-world scenarios and 7 categories of real malicious websites. Experiments with 12 popular LLMs reveal that existing models struggle to detect elaborately disguised malicious URLs. We further identify and analyze key factors that impact attack success rates and propose URLGuard, a lightweight defense module. We believe this work will provide a foundational resource for advancing the security of web agents. Our code is available at https://github.com/JiangYingEr/MalURLBench.

Large Language Models (LLMs) have emerged as a transformative and disruptive technology, enabling a wide range of applications in natural language processing, machine translation, and beyond. However, this widespread integration of LLMs also raised several security concerns highlighted by the Open Web Application Security Project (OWASP), which has identified the top 10 security vulnerabilities inherent in LLM applications. Addressing these vulnerabilities is crucial, given the increasing reliance on LLMs and the potential threats to data integrity, confidentiality, and service availability. This paper presents a framework designed to mitigate the security risks outlined in the OWASP Top 10. Our proposed model leverages LLM-enabled intelligent agents, offering a new approach to proactively identify, assess, and counteract security threats in real-time. The proposed framework serves as an initial blueprint for future research and development, aiming to enhance the security measures of LLMs and protect against emerging threats in this rapidly evolving landscape.

Aim-assist cheats are the most prevalent and infamous form of cheating in First-Person Shooter (FPS) games, which help cheaters illegally reveal the opponent's location and auto-aim and shoot, and thereby pose significant threats to the game industry. Although a considerable research effort has been made to automatically detect aim-assist cheats, existing works suffer from unreliable frameworks, limited generalizability, high overhead, low detection performance, and a lack of explainability of detection results. In this paper, we propose XGuardian, a server-side generalized and explainable system for detecting aim-assist cheats to overcome these limitations. It requires only two raw data inputs, pitch and yaw, which are all FPS games' must-haves, to construct novel temporal features and describe aim trajectories, which are essential for distinguishing cheaters and normal players. XGuardian is evaluated with the latest mainstream FPS game CS2, and validates its generalizability with another two different games. It achieves high detection performance and low overhead compared to prior works across different games with real-world and large-scale datasets, demonstrating wide generalizability and high effectiveness. It is able to justify its predictions and thereby shorten the ban cycle. We make XGuardian as well as our datasets publicly available.

We show reductions and equivalences between various problems related to the computation of the endomorphism ring of principally polarised superspecial abelian surfaces. Problems considered are the computation of the Ibukiyama-Katsura-Oort matrix and computation of unpolarised isomoprhisms between superspecial abelian surfaces.

Aggregate analytics over conversational data are increasingly used for safety monitoring, governance, and product analysis in large language model systems. A common practice is to embed conversations, cluster them, and publish short textual summaries describing each cluster. While raw conversations may never be exposed, these derived summaries can still pose privacy risks if they contain personally identifying information (PII) or uniquely traceable strings copied from individual conversations. We introduce CanaryBench, a simple and reproducible stress test for privacy leakage in cluster-level conversation summaries. CanaryBench generates synthetic conversations with planted secret strings ("canaries") that simulate sensitive identifiers. Because canaries are known a priori, any appearance of these strings in published summaries constitutes a measurable leak. Using TF-IDF embeddings and k-means clustering on 3,000 synthetic conversations (24 topics) with a canary injection rate of 0.60, we evaluate an intentionally extractive example snippet summarizer that models quote-like reporting. In this configuration, we observe canary leakage in 50 of 52 canary-containing clusters (cluster-level leakage rate 0.961538), along with nonzero regex-based PII indicator counts. A minimal defense combining a minimum cluster-size publication threshold (k-min = 25) and regex-based redaction eliminates measured canary leakage and PII indicator hits in the reported run while maintaining a similar cluster-coherence proxy. We position this work as a societal impacts contribution centered on privacy risk measurement for published analytics artifacts rather than raw user data.

Auditability and reproducibility still are critical challenges for real-time data streams pipelines. Streaming engines are highly dependent on runtime scheduling, window triggers, arrival orders, and uncertainties such as network jitters. These all derive the streaming pipeline platforms to throw non-determinist outputs. In this work, we introduce a blockchain-backed provenance architecture for streaming platform (e.g Kafka Streams) the publishes cryptographic data of a windowed data stream without publishing window payloads on-chain. We used real-time weather data from weather stations in Berlin. Weather records are canonicalized, deduplicated, and aggregated per window, then serialised deterministically. Furthermore, the Merkle root of the records within the window is computed and stored alongside with Kafka offsets boundaries to MultiChain blockchain streams as checkpoints. Our design can enable an independent auditor to verify: (1) the completeness of window payloads, (2) canonical serialization, and (3) correctness of derived analytics such as minimum/maximum/average temperatures. We evaluated our system using real data stream from two weather stations (Berlin-Brandenburg and Berlin-Tempelhof) and showed linear verification cost, deterministic reproducibility, and with a scalable off-chain storage with on-chain cryptographic anchoring. We also demonstrated that the blockchain can afford to be integrated with streaming platforms particularly with our system, and we get satisfactory transactions per second values.

Data at the physical layer transmits via media such as copper cable, fiber optic, or wireless. Physical attack vectors exist that challenge data confidentiality and availability. Protocols and encryption standards help obfuscate but often cannot keep the data type and destination secure, with limited insight into confidentiality and integrity. We will investigate the feasibility of developing an awareness and integrity protocol to help mitigate physical side-channel attacks that lead to eavesdropping of data communication and denial-of-service. Keywords: data confidentiality, siphoning, eavesdropping, person-in-the-middle, denial-of-service, physical layer attacks, nation-states