Papers

This paper presents LLMBugScanner, a large language model (LLM) based framework for smart contract vulnerability detection using fine-tuning and ensemble learning. Smart contract auditing presents several challenges for LLMs: different pretrained models exhibit varying reasoning abilities, and no single model performs consistently well across all vulnerability types or contract structures. These limitations persist even after fine-tuning individual LLMs. To address these challenges, LLMBugScanner combines domain knowledge adaptation with ensemble reasoning to improve robustness and generalization. Through domain knowledge adaptation, we fine-tune LLMs on complementary datasets to capture both general code semantics and instruction-guided vulnerability reasoning, using parameter-efficient tuning to reduce computational cost. Through ensemble reasoning, we leverage the complementary strengths of multiple LLMs and apply a consensus-based conflict resolution strategy to produce more reliable vulnerability assessments. We conduct extensive experiments across multiple popular LLMs and compare LLMBugScanner with both pretrained and fine-tuned individual models. Results show that LLMBugScanner achieves consistent accuracy improvements and stronger generalization, demonstrating that it provides a principled, cost-effective, and extensible framework for smart contract auditing.

Deep learning models are used in safety-critical tasks such as automated driving and face recognition. However, small perturbations in the model input can significantly change the predictions. Adversarial attacks are used to identify small perturbations that can lead to misclassifications. More powerful black-box adversarial attacks are required to develop more effective defenses. A promising approach to black-box adversarial attacks is to repeat the process of extracting a specific image area and changing the perturbations added to it. Existing attacks adopt simple rectangles as the areas where perturbations are changed in a single iteration. We propose applying superpixels instead, which achieve a good balance between color variance and compactness. We also propose a new search method, versatile search, and a novel attack method, Superpixel Attack, which applies superpixels and performs versatile search. Superpixel Attack improves attack success rates by an average of 2.10% compared with existing attacks. Most models used in this study are robust against adversarial attacks, and this improvement is significant for black-box adversarial attacks. The code is avilable at https://github.com/oe1307/SuperpixelAttack.git.

Digital forensics faces unprecedented challenges with the emergence of digital twins and metaverse technologies. This paper presents the first comparative analysis between blockchain-based and traditional database systems for managing digital twin evidence in forensic investigations. We conducted controlled experiments comparing the Ethereum blockchain with IPFS storage against traditional SQL databases for digital twin evidence management. Our findings reveal that while blockchain provides superior data integrity and immutability, crucial for forensic applications, traditional databases offer better performance consistency. The blockchain implementation showed faster average storage times but higher variability in retrieval operations. Both systems maintained forensic integrity through hash verification, though blockchain's immutable nature provides additional security guarantees essential for legal proceedings. This research contributes to the development of robust digital forensic methodologies for emerging technologies in the metaverse era.

The gamut of todays internet-connected embedded devices has led to increased concerns regarding the security and confidentiality of data. Most internet-connected embedded devices employ mathematically secure cryptographic algorithms to address security vulnerabilities. Despite such mathematical guarantees, as these algorithms are often implemented in silicon, they leak critical information in terms of power consumption, electromagnetic (EM) radiation, timing, cache hits and misses, photonic emission and so on, leading to side-channel analysis (SCA) attacks. This thesis focuses on low overhead generic circuit-level yet synthesizable countermeasures against power and EM SCA. Existing countermeasures (including proposed) still have relatively high overhead which bars them from being used in energy-constraint IoT devices. We propose a zero-overhead integrated inductive sensor which is able to detect i)EM SCA ii) Clock glitch-based Fault Injection Attack (FIA), and iii) Voltage-glitch based Fault Injection Attack by using a simple ML algorithm. Advent of quantum computer research will open new possibilities for theoretical attacks against existing cryptographic protocols. National Institute of Standard & Technology (NIST) has standardized post-quantum cryptographic algorithms to secure crypto-systems against quantum adversary. I contribute to the standardization procedure by introducing the first silicon-verified Saber (a NIST finalist modulo Learning with Rounding scheme) which consumes lowest energy and area till date amongst all the candidates.

Modern AI inference faces an irreducible tension: no single computational resource simultaneously maximizes performance, preserves privacy, minimizes cost, and maintains trust. Existing orchestration frameworks optimize single dimensions (Kubernetes prioritizes latency, federated learning preserves privacy, edge computing reduces network distance), creating solutions that struggle under real-world heterogeneity. We present IslandRun, a multi-objective orchestration system that treats computational resources as autonomous "islands" spanning personal devices, private edge servers, and public cloud. Our key insights: (1) request-level heterogeneity demands policy-constrained multi-objective optimization, (2) data locality enables routing compute to data rather than data to compute, and (3) typed placeholder sanitization preserves context semantics across trust boundaries. IslandRun introduces agent-based routing, tiered island groups with differential trust, and reversible anonymization. This establishes a new paradigm for privacy-aware, decentralized inference orchestration across heterogeneous personal computing ecosystems.

Hardware Trojans (HT s) are a persistent threat to integrated circuits, especially when inserted at the register-transfer level (RTL). Existing methods typically first convert the design into a graph, such as a gate-level netlist or an RTL-derived dataflow graph (DFG), and then use a graph neural network (GNN ) to obtain an embedding of that graph, which (i) loses compact RTL semantics, (ii) relies on shallow GNNs with limited receptive field, and (iii) is largely restricted to coarse, module-level binary HT detection. We propose TrojanLoC, an LLM-based framework for RTL-level HT localization. We use an RTL-finetuned LLM to derive module-level and line-level embeddings directly from RTL code, capturing both global design context and local semantics. Next, we train task-specific classifiers on these embeddings to perform module-level Trojan detection, type prediction, and fine-grained line-level localization. We also introduce TrojanInS, a large synthetic dataset of RTL designs with systematically injected Trojans from four effect-based categories, each accompanied by precise line-level annotations. Our experiments show that TrojanLoC achieves strong module-level performance, reaching 0.99 F1-score for Trojan detection, up to 0.68 higher than baseline, and 0.84 macro-F1 for Trojan-type classification. At the line level, TrojanLoc further achieves up to 0.93 macro-F1, enabling fine-grained localization of Trojan-relevant RTL lines

Retrieving up-to-date information from a publicly accessible database poses significant threats to the user's privacy. Private information retrieval (PIR) protocols allow a user to retrieve any entry from a database, without revealing the identity of the entry being retrieved to the server(s). Such protocols have found numerous applications in both theoretical studies and real-life scenarios. The existing PIR constructions mainly give multi-server information-theoretic PIR (IT-PIR) protocols or single-server computational PIR (CPIR) protocols. Compared with CPIR, IT-PIR protocols are computationally more efficient and secure in the presence of unbounded servers. The most classical and challenging problem in the realm of IT-PIR is constructing protocols with lower communication complexity. In this review, we introduce a new discrete structure called families of orthogonal arrays with span capability (FOASC) and propose a unified framework for constructing IT-PIR protocols. We show how the most influential IT-PIR protocols in the literature can be captured by the framework. We also put forward several interesting open problems concerning FOASC, whose solutions may result in innovative IT-PIR protocols.

Tor is a widely used anonymity network that conceals user identities by routing traffic through encrypted relays, yet it remains vulnerable to traffic correlation attacks that deanonymize users by matching patterns in ingress and egress traffic. However, existing correlation methods suffer from two major limitations: limited robustness to noise and partial observations, and poor scalability due to computationally expensive pairwise matching. To address these challenges, we propose RECTor, a machine learning-based framework for traffic correlation under realistic conditions. RECTor employs attention-based Multiple Instance Learning (MIL) and GRU-based temporal encoding to extract robust flow representations, even when traffic data is incomplete or obfuscated. These embeddings are mapped into a shared space via a Siamese network and efficiently matched using approximate nearest neighbor (aNN) search. Empirical evaluations show that RECTor outperforms state-of-the-art baselines such as DeepCorr, DeepCOFFEA, and FlowTracker, achieving up to 60% higher true positive rates under high-noise conditions and reducing training and inference time by over 50%. Moreover, RECTor demonstrates strong scalability: inference cost grows near-linearly as the number of flows increases. These findings reveal critical vulnerabilities in Tor's anonymity model and highlight the need for advanced model-aware defenses.

Data privacy is a critical challenge in modern medical workflows as the adoption of electronic patient records has grown rapidly. Stringent data protection regulations limit access to clinical records for training and integrating machine learning models that have shown promise in improving diagnostic accuracy and personalized care outcomes. Synthetic data offers a promising alternative; however, current generative models either struggle with time-series data or lack formal privacy guaranties. In this paper, we enhance a state-of-the-art time-series generative model to better handle longitudinal clinical data while incorporating quantifiable privacy safeguards. Using real data from chronic kidney disease and ICU patients, we evaluate our method through statistical tests, a Train-on-Synthetic-Test-on-Real (TSTR) setup, and expert clinical review. Our non-private model (Augmented TimeGAN) outperforms transformer- and flow-based models on statistical metrics in several datasets, while our private model (DP-TimeGAN) maintains a mean authenticity of 0.778 on the CKD dataset, outperforming existing state-of-the-art models on the privacy-utility frontier. Both models achieve performance comparable to real data in clinician evaluations, providing robust input data necessary for developing models for complex chronic conditions without compromising data privacy.

This paper introduces BeaCon, a novel tool for the automated generation of adjustable container security policies. Unlike prior approaches, BeaCon leverages dynamic analysis to simulate realistic environments, uncovering container execution paths that may remain hidden during the profiling phase. To address the challenge of exploring vast profiling spaces, we employ efficient heuristics to reveal additional system events with minimal effort. In addition, BeaCon incorporates a security and functionality scoring mechanism to prioritize system calls and capabilities based on their impact on the host OS kernel's security and the functionality of containerized applications. By integrating these scores, BeaCon achieves a customized balance between security and functionality, enabling cloud providers to enforce security measures while maintaining tenant availability. We implemented a prototype of BeaCon using eBPF kernel technology and conducted extensive evaluations. Results from the top 15 containers, which revealed significant improvements, demonstrate that BeaCon identifies an average of 16.5% additional syscalls by applying diverse environments. Furthermore, we evaluated its effectiveness in mitigating risks associated with 45 known vulnerabilities (e.g., CVEs), showcasing its potential to significantly enhance container security. Additionally, we performed proof-of-concept demonstrations for two well-known security vulnerabilities, showing that BeaCon successfully reduces attack surface by blocking these exploits.

Large Reasoning Models (LRMs) have emerged as a powerful advancement in multi-step reasoning tasks, offering enhanced transparency and logical consistency through explicit chains of thought (CoT). However, these models introduce novel safety and reliability risks, such as CoT-hijacking and prompt-induced inefficiencies, which are not fully captured by existing evaluation methods. To address this gap, we propose RT-LRM, a unified benchmark designed to assess the trustworthiness of LRMs. RT-LRM evaluates three core dimensions: truthfulness, safety and efficiency. Beyond metric-based evaluation, we further introduce the training paradigm as a key analytical perspective to investigate the systematic impact of different training strategies on model trustworthiness. We achieve this by designing a curated suite of 30 reasoning tasks from an observational standpoint. We conduct extensive experiments on 26 models and identify several valuable insights into the trustworthiness of LRMs. For example, LRMs generally face trustworthiness challenges and tend to be more fragile than Large Language Models (LLMs) when encountering reasoning-induced risks. These findings uncover previously underexplored vulnerabilities and highlight the need for more targeted evaluations. In addition, we release a scalable toolbox for standardized trustworthiness research to support future advancements in this important field. Our code and datasets will be open-sourced.

Memecoins, emerging from internet culture and community-driven narratives, have rapidly evolved into a unique class of crypto assets. Unlike technology-driven cryptocurrencies, their market dynamics are primarily shaped by viral social media diffusion, celebrity influence, and speculative capital inflows. To capture the distinctive vulnerabilities of these ecosystems, we present the first Memecoin Ecosystem Fragility Framework (ME2F). ME2F formalizes memecoin risks in three dimensions: i) Volatility Dynamics Score capturing persistent and extreme price swings together with spillover from base chains; ii) Whale Dominance Score quantifying ownership concentration among top holders; and iii) Sentiment Amplification Score measuring the impact of attention-driven shocks on market stability. We apply ME2F to representative tokens (over 65% market share) and show that fragility is not evenly distributed across the ecosystem. Politically themed tokens such as TRUMP, MELANIA, and LIBRA concentrate the highest risks, combining volatility, ownership concentration, and sensitivity to sentiment shocks. Established memecoins such as DOGE, SHIB, and PEPE fall into an intermediate range. Benchmark tokens ETH and SOL remain consistently resilient due to deeper liquidity and institutional participation. Our findings provide the first ecosystem-level evidence of memecoin fragility and highlight governance implications for enhancing market resilience in the Web3 era.

Approximate machine unlearning aims to efficiently remove the influence of specific data points from a trained model, offering a practical alternative to full retraining. However, it introduces privacy risks: an adversary with access to pre- and post-unlearning models can exploit their differences for membership inference or data reconstruction. We show these vulnerabilities arise from two factors: large gradient norms of forget-set samples and the close proximity of unlearned parameters to the original model. To demonstrate their severity, we propose unlearning-specific membership inference and reconstruction attacks, showing that several state-of-the-art methods (e.g., NGP, SCRUB) remain vulnerable. To mitigate this leakage, we introduce WARP, a plug-and-play teleportation defense that leverages neural network symmetries to reduce forget-set gradient energy and increase parameter dispersion while preserving predictions. This reparameterization obfuscates the signal of forgotten data, making it harder for attackers to distinguish forgotten samples from non-members or recover them via reconstruction. Across six unlearning algorithms, our approach achieves consistent privacy gains, reducing adversarial advantage (AUC) by up to 64% in black-box and 92% in white-box settings, while maintaining accuracy on retained data. These results highlight teleportation as a general tool for reducing attack success in approximate unlearning.

The growing presence of IoT devices in academic environments has increased operational complexity and exposed security weaknesses, especially in academic institutions without unified policies for registration, monitoring, and incident response involving IoT. This work presents IoTEdu, an integrated platform that combines access control, incident detection, and automatic blocking of IoT devices. The solution was evaluated in a controlled environment with simulated attacks, achieving an average time of 28.6 seconds between detection and blocking. The results show a reduction in manual intervention, standardization of responses, and unification of the processes of registration, monitoring, and incident response.

Few-shot prompting has emerged as a practical alternative to fine-tuning for leveraging the capabilities of large language models (LLMs) in specialized tasks. However, its effectiveness depends heavily on the selection and quality of in-context examples, particularly in complex domains. In this work, we examine retrieval-augmented prompting as a strategy to improve few-shot performance in code vulnerability detection, where the goal is to identify one or more security-relevant weaknesses present in a given code snippet from a predefined set of vulnerability categories. We perform a systematic evaluation using the Gemini-1.5-Flash model across three approaches: (1) standard few-shot prompting with randomly selected examples, (2) retrieval-augmented prompting using semantically similar examples, and (3) retrieval-based labeling, which assigns labels based on retrieved examples without model inference. Our results show that retrieval-augmented prompting consistently outperforms the other prompting strategies. At 20 shots, it achieves an F1 score of 74.05% and a partial match accuracy of 83.90%. We further compare this approach against zero-shot prompting and several fine-tuned models, including Gemini-1.5-Flash and smaller open-source models such as DistilBERT, DistilGPT2, and CodeBERT. Retrieval-augmented prompting outperforms both zero-shot (F1 score: 36.35%, partial match accuracy: 20.30%) and fine-tuned Gemini (F1 score: 59.31%, partial match accuracy: 53.10%), while avoiding the training time and cost associated with model fine-tuning. On the other hand, fine-tuning CodeBERT yields higher performance (F1 score: 91.22%, partial match accuracy: 91.30%) but requires additional training, maintenance effort, and resources.

Large Language Models (LLMs) have demonstrated exceptional capabilities in natural language understanding and generation. Based on these LLMs, businesses have started to provide Embeddings-as-a-Service (EaaS), offering feature extraction capabilities (in the form of text embeddings) that benefit downstream natural language processing tasks. However, prior research has demonstrated that EaaS is vulnerable to imitation attacks, where an attacker clones the service's model in a black-box manner without access to the model's internal workings. In response, watermarks have been added to the text embeddings to protect the intellectual property of EaaS providers by allowing them to check for model ownership. This thesis focuses on defending against imitation attacks by investigating EaaS watermarks. To achieve this goal, we unveil novel attacks and propose and validate new watermarking techniques. Firstly, we show that existing EaaS watermarks can be removed through paraphrasing the input text when attackers clone the model during imitation attacks. Our study illustrates that paraphrasing can effectively bypass current state-of-the-art EaaS watermarks across various attack setups (including different paraphrasing techniques and models) and datasets in most instances. This demonstrates a new vulnerability in recent EaaS watermarking techniques. Subsequently, as a countermeasure, we propose a novel watermarking technique, WET (Watermarking EaaS with Linear Transformation), which employs linear transformation of the embeddings. Watermark verification is conducted by applying a reverse transformation and comparing the similarity between recovered and original embeddings. We demonstrate its robustness against paraphrasing attacks with near-perfect verifiability. We conduct detailed ablation studies to assess the significance of each component and hyperparameter in WET.

The increasing complexity of IoT edge networks presents significant challenges for anomaly detection, particularly in identifying sophisticated Denial-of-Service (DoS) attacks and zero-day exploits under highly dynamic and imbalanced traffic conditions. This paper proposes SD-CGAN, a Conditional Generative Adversarial Network framework enhanced with Sinkhorn Divergence, tailored for robust anomaly detection in IoT edge environments. The framework incorporates CTGAN-based synthetic data augmentation to address class imbalance and leverages Sinkhorn Divergence as a geometry-aware loss function to improve training stability and reduce mode collapse. The model is evaluated on exploitative attack subsets from the CICDDoS2019 dataset and compared against baseline deep learning and GAN-based approaches. Results show that SD-CGAN achieves superior detection accuracy, precision, recall, and F1-score while maintaining computational efficiency suitable for deployment in edge-enabled IoT environments.

AI systems that output their reasoning in natural language offer an opportunity for safety – we can monitor their chain of thought (CoT) for undesirable reasoning, such as the pursuit of harmful objectives. However, the extent to which CoT faithfully reflects the underlying reasoning process, and hence the extent to which it can be usefully monitored, may be influenced by certain aspects of training. We investigate how different training incentives, applied to a reasoning model, affect its monitorability. We introduce a novel methodology for measuring monitorability according to whether a monitor can predict a key latent variable using the model's reasoning. When controlling for accuracy, we do not find evidence for consistent effects from commonly used incentives (length penalties and KL regularisation), but we find that adversarial optimisation (penalising monitor accuracy) degrades monitor performance, while direct optimisation for monitorability does not reliably lead to improvements. Our code is available at https://github.com/QiyaoWei/reasoning-under-pressure.

This research introduces the Decentralized Finance (DeFi) TrustBoost Framework, which combines blockchain technology and Explainable AI to address challenges faced by lenders underwriting small business loan applications from low-wealth households. The framework is designed with a strong emphasis on fulfilling four crucial requirements of blockchain and AI systems: confidentiality, compliance with data protection laws, resistance to adversarial attacks, and compliance with regulatory audits. It presents a technique for tamper-proof auditing of automated AI decisions and a strategy for on-chain (inside-blockchain) and off-chain data storage to facilitate collaboration within and across financial organizations.

Equipped with various tools and knowledge, GPTs, one kind of customized AI agents based on OpenAI's large language models, have illustrated great potential in many fields, such as writing, research, and programming. Today, the number of GPTs has reached three millions, with the range of specific expert domains becoming increasingly diverse. However, given the consistent framework shared among these LLM agent applications, systemic security vulnerabilities may exist and remain underexplored. To fill this gap, we present an empirical study on the security vulnerabilities of GPTs. Building upon prior research on LLM security, we first adopt a platform-user perspective to conduct a comprehensive attack surface analysis across different system components. Then, we design a systematic and multidimensional attack suite with the explicit objectives of information leakage and tool misuse based on the attack surface analysis, thereby concretely demonstrating the security vulnerabilities that various components of GPT-based systems face. Finally, we accordingly propose defense mechanisms to address the aforementioned security vulnerabilities. By increasing the awareness of these vulnerabilities and offering critical insights into their implications, this study seeks to facilitate the secure and responsible application of GPTs while contributing to developing robust defense mechanisms that protect users and systems against malicious attacks.

Automated vulnerability patching is crucial for software security, and recent advancements in Large Language Models (LLMs) present promising capabilities for automating this task. However, existing research has primarily assessed LLMs using publicly disclosed vulnerabilities, leaving their effectiveness on related artificial vulnerabilities largely unexplored. In this study, we empirically evaluate the patching effectiveness and complementarity of several prominent LLMs, such as OpenAI's GPT variants, LLaMA, DeepSeek, and Mistral models, using both real and artificial vulnerabilities. Our evaluation employs Proof-of-Vulnerability (PoV) test execution to concretely assess whether LLM-generated source code successfully patches vulnerabilities. Our results reveal that LLMs patch real vulnerabilities more effectively compared to artificial ones. Additionally, our analysis reveals significant variability across LLMs in terms of overlapping (multiple LLMs patching the same vulnerabilities) and complementarity (vulnerabilities patched exclusively by a single LLM), emphasizing the importance of selecting appropriate LLMs for effective vulnerability patching.

In this paper, we explore how quantum resources can be used to increase the rate of private distributed matrix multiplication (PDMM). In PDMM, a user who has two high-dimensional matrices, A and B, and lacks the computational capabilities to apply matrix multiplication locally, divides the matrices A and B into K and L sub-blocks, respectively. Then, the user sends them to N servers to apply the required multiplication privately, i.e., any $T$ colluding servers cannot get any information about the user's matrices. The goal is to reduce the number of servers needed to perform the required matrix multiplication, thereby decreasing the communication cost. First, in the high-privacy regime, the state-of-the-art classical code is called the gap additive secure polynomial (GASP) code. We define a feasibility requirement in the quantum setting for the GASP code such that the highest performance is achieved when the requirement is satisfied. Thus, super-dense coding gain is achieved when the feasibility condition is satisfied. We show that when $T \geq KL-K+1$, the feasibility condition is always satisfied and the GASP code can be extended to the quantum version. In the case of $T < KL-K+1$, the feasibility can still be satisfied. To further examine this behavior, we numerically study how the minimum privacy requirement depends on the matrix dimensions and provide a quadratic estimate for this relation. The results suggest that feasibility can be achieved when $T \sim 0.5 KL$. Second, in the low-privacy regime, the recently developed cyclic-addition degree tables (CAT) and discretely optimized GASP (DOG) codes are among the most efficient known classical constructions for PDMM. We show that the feasibility condition developed for GASP can be adopted for both CAT and DOG codes as well, thus unifying the feasibility framework for multiple classical PDMM coding schemes.

Federated Learning (FL) enables collaborative, privacy-preserving model training, but supporting the "Right to be Forgotten" is especially challenging because data influences the model through distributed and interleaved client updates. Existing exact unlearning methods typically require frequent retraining from scratch, resulting in high communication cost and long service downtime. To address this, we propose Federated Sequential Group-based Training (FedSGT), an exact unlearning framework for FL. FedSGT partitions the data into uniform groups, and each client may participate in multiple groups. To control communication overhead, each client can limit the number of groups it contributes to. FedSGT then trains multiple sequences of Parameter-Efficient Fine-Tuning (PEFT) modules, each corresponding to a different group permutation. Since the PEFT modules are lightweight and maintained server-side, FedSGT isolates the influence of different data groups into independent modules without incurring significant storage overhead and communication cost. Exact unlearning is thus achieved instantly by deactivating the modules corresponding to the group containing the unlearned data. Furthermore, using multiple training sequences helps maintain high model utility as deletion requests accumulate. We provide a rigorous theoretical analysis of both the deletion rate – expected number of deletions before retraining is needed – and the expected model performance. Experiments on various tasks demonstrate that FedSGT achieves a significantly longer service maintenance under multiple unlearning requests while maintaining comparable learning performance and training efficiency to other exact unlearning baselines. Extensive ablation studies validate the robustness of our method across a wide range of parameter settings.

Modern cloud applications are built from independent microservices, offering scalability and usage-based billing. However, their reliance on independently-operating auto-scalers introduces coordination challenges. Default retry patterns can trigger "retry storms" during service miscoordination or adversarial overload, amplifying load, latency, and resource billing. These storms can cause either self-inflicted Denial-of-Wallet (DoW) or amplify the impact of DDoS attacks. To overcome these problems, we introduce RetryGuard, a distributed framework for productive control of retry patterns across interdependent microservices. By managing retry policy on a per-service basis and making parallel decisions, RetryGuard prevents retry storms, curbs resource contention, and mitigates escalating operational costs. RetryGuard makes its decisions based on an analytical model that captures the relationships among retries, throughput (rejections), delays, and costs. Simulations show that it outperforms established mechanisms, including exponential backoff, jitter, and retry budgets. Experimental results show that RetryGuard significantly reduces resource usage and costs compared to existing AWS policies, achieving more than 90% improvement in latency, and 98% reduction in storm size. We further demonstrate its effectiveness in a multi-layer Kubernetes deployment with the Istio service-mesh, where RetryGuard reduces the peak number of replicas by $3 \times$ and cumulative memory usage by 55%.

Federated Learning (FL) offers a promising approach to collaboratively train machine learning models without centralizing raw data, yet its scalability is often throttled by excessive communication overhead. This challenge is magnified in Internet of Things (IoT) environments, where devices face stringent bandwidth, latency, and energy constraints. Conventional secure aggregation protocols, while essential for protecting model updates, frequently require multiple interaction rounds, large payload sizes, and per-client costs rendering them impractical for many edge deployments. In this work, we present Hyb-Agg, a lightweight and communication-efficient secure aggregation protocol that integrates Multi-Key CKKS (MK-CKKS) homomorphic encryption with Elliptic Curve Diffie-Hellman (ECDH)-based additive masking. Hyb-Agg reduces the secure aggregation process to a single, non-interactive client-to-server transmission per round, ensuring that per-client communication remains constant regardless of the number of participants. This design eliminates partial decryption exchanges, preserves strong privacy under the RLWE, CDH, and random oracle assumptions, and maintains robustness against collusion by the server and up to $N-2$ clients. We implement and evaluate Hyb-Agg on both high-performance and resource-constrained devices, including a Raspberry Pi 4, demonstrating that it delivers sub-second execution times while achieving a constant communication expansion factor of approximately 12x over plaintext size. By directly addressing the communication bottleneck, Hyb-Agg enables scalable, privacy-preserving federated learning that is practical for real-world IoT deployments.

Psychological stress is a widespread issue that significantly impacts student well-being and academic performance. Effective remote stress recognition is crucial, yet existing methods often rely on wearable devices or GPS-based clustering techniques that pose privacy risks and lack of human understandable explanations. In this study, we introduce a novel, end-to-end privacy-enhanced framework for semantic location encoding using a self-hosted OSM engine and an LLM-bootstrapped static map for human-friendly feature extraction, and pave a pathway for privacy-aware location data transformation for dataset sharing. We rigorously quantify the privacy-utility-explainability trilemma and demonstrate (via LOSO validation) that our Privacy-Aware (PA) model achieves robust privacy protection without being statistically distinguishable in stress recognition performance from a non-private model. Model explanation analysis highlights that our extracted features, which are user-friendly features, match with psychological literature about stress. In addition, an ablation study on the GeoLife dataset also demonstrates that our privacy framework improves privacy by 2-3 times compared to a non-privacy-aware approach. This suggests that our system can be utilized for the next generation of GPS transformations in open-source datasets for future researchers.

Recent years have shown that malware attacks still happen with high frequency. Malware experts seek to categorize and classify incoming samples to confirm their trustworthiness or prove their maliciousness. One of the ways in which groups of malware samples can be identified is through malware clustering. Despite the efforts of the community, malware clustering which incorporates benign samples has been under-explored. Moreover, despite the availability of larger public benchmark malware datasets, malware clustering studies have avoided fully utilizing these datasets in their experiments, often resorting to small datasets with only a few families. Additionally, the current state-of-the-art solutions for malware clustering remain unclear. In our study, we evaluate malware clustering quality and establish the state-of-the-art on Bodmas and Ember - two large public benchmark malware datasets. Ours is the first study of malware clustering performed on whole malware benchmark datasets. Additionally, we extend the malware clustering task by incorporating benign samples. Our results indicate that incorporating benign samples does not significantly degrade clustering quality. We find that there are differences in the quality of the created clusters between Ember and Bodmas, as well as a private industry dataset. Contrary to popular opinion, our top clustering performers are K-Means and BIRCH, with DBSCAN and HAC falling behind.

Given the constant growth and increasing sophistication of cyberattacks, cybersecurity can no longer rely solely on traditional defense techniques and tools. Proactive detection of cyber threats has become essential to help security teams identify potential risks and implement effective mitigation measures. Cyber Threat Intelligence (CTI) plays a key role by providing security analysts with evidence-based knowledge about cyber threats. CTI information can be extracted using various techniques and data sources; however, machine learning has proven promising. As for data sources, social networks and online discussion forums are commonly explored. In this study, we apply text mining techniques and machine learning to data collected from Dark Web forums in Brazilian Portuguese to identify malicious posts. Our contributions include the creation of three original datasets, a novel multi-stage labeling process combining indicators of compromise (IoCs), contextual keywords, and manual analysis, and a comprehensive evaluation of text representations and classifiers. To our knowledge, this is the first study to focus specifically on Brazilian Portuguese content in this domain. The best-performing model, using LightGBM and TF-IDF, was able to detect relevant posts with high accuracy. We also applied topic modeling to validate the model's outputs on unlabeled data, confirming its robustness in real-world scenarios.

Every day we share our personal information through digital systems which are constantly exposed to threats. For this reason, security-oriented disciplines of signal processing have received increasing attention in the last decades: multimedia forensics, digital watermarking, biometrics, network monitoring, steganography and steganalysis are just a few examples. Even though each of these fields has its own peculiarities, they all have to deal with a common problem: the presence of one or more adversaries aiming at making the system fail. Adversarial Signal Processing lays the basis of a general theory that takes into account the impact that the presence of an adversary has on the design of effective signal processing tools. By focusing on the application side of Adversarial Signal Processing, namely adversarial information fusion in distributed sensor networks, and adopting a game-theoretic approach, this thesis contributes to the above mission by addressing four issues. First, we address decision fusion in distributed sensor networks by developing a novel soft isolation defense scheme that protect the network from adversaries, specifically, Byzantines. Second, we develop an optimum decision fusion strategy in the presence of Byzantines. In the next step, we propose a technique to reduce the complexity of the optimum fusion by relying on a novel near-optimum message passing algorithm based on factor graphs. Finally, we introduce a defense mechanism to protect decentralized networks running consensus algorithm against data falsification attacks.

Large Language Model (LLM)-based Multi-Agent Systems (MAS) are susceptible to linguistic attacks that can trigger cascading failures across the network. Existing defenses face a fundamental dilemma: lightweight single-auditor methods are prone to single points of failure, while robust committee-based approaches incur prohibitive computational costs in multi-turn interactions. To address this challenge, we propose MAS-Shield, a secure and efficient defense framework designed with a coarse-to-fine filtering pipeline. Rather than applying uniform scrutiny, MAS-Shield dynamically allocates defense resources through a three-stage protocol: (1) Critical Agent Selection strategically targets high-influence nodes to narrow the defense surface; (2) Light Auditing employs lightweight sentry models to rapidly filter the majority of benign cases; and (3) Global Consensus Auditing escalates only suspicious or ambiguous signals to a heavyweight committee for definitive arbitration. This hierarchical design effectively optimizes the security-efficiency trade-off. Experiments demonstrate that MAS-Shield achieves a 92.5% recovery rate against diverse adversarial scenarios and reduces defense latency by over 70% compared to existing methods.