FedShield-LLM: a Secure and Scalable Federated Fine-Tuned Large Language Model

Federated Learning FL offers a decentralized framework for training and fine-tuning Large Language Models LLMs by leveraging computational resources across organizations while keeping sensitive data on local devices. It addresses privacy and security concerns while navigating challenges associate...

FERRET: Private Deep Learning Faster and Better Than DPSGD

We revisit 1-bit gradient compression through the lens of mutual-information differential privacy MI-DP. Building on signSGD, we propose FERRET--Fast and Effective Restricted Release for Ethical Training--which transmits at most one sign bit per parameter group with Bernoulli masking. Theory: We...

Heterogeneous Secure Transmissions in IRS-Assisted NOMA Communications: CO-GNN Approach

Intelligent Reflecting Surfaces IRS enhance spectral efficiency by adjusting reflection phase shifts, while Non-Orthogonal Multiple Access NOMA increases system capacity. Consequently, IRS-assisted NOMA communications have garnered significant research interest. However, the passive nature of the...

Sylva: Tailoring Personalized Adversarial Defense in Pre-Trained Models Via Collaborative Fine-Tuning

Whitepaper called Sylva: Tailoring Personalized Adversarial Defense In Pre-Trained Models Via Collaborative Fine-Tuning...

Request-Baskets Server-Side Request Forgery

Request-Baskets versions up to 1.2.1 proof of concept server-side request forgery exploit...

SAP GuiXT Scripting Issues

Multiple vulnerabilities have been discovered in SAP GuiXT scripting, which could allow an attacker to perform remote code execution, steal NTLM hashes, conduct client-side request forgery attacks, and launch denial of service DoS attacks. These vulnerabilities arise from insecure design principl...

COALESCE: Economic and Security Dynamics of Skill-Based Task Outsourcing among Team of Autonomous LLM Agents

The meteoric rise and proliferation of autonomous Large Language Model LLM agents promise significant capabilities across various domains. However, their deployment is increasingly constrained by substantial computational demands, specifically for Graphics Processing Unit GPU resources. This pape...

Privacy-Aware, Public-Aligned: Embedding Risk Detection and Public Values into Scalable Clinical Text De-Identification for Trusted Research Environments

Clinical free-text data offers immense potential to improve population health research such as richer phenotyping, symptom tracking, and contextual understanding of patient care. However, these data present significant privacy risks due to the presence of directly or indirectly identifying...

SpeechVerifier: Robust Acoustic Fingerprint against Tampering Attacks Via Watermarking

With the surge of social media, maliciously tampered public speeches, especially those from influential figures, have seriously affected social stability and public trust. Existing speech tampering detection methods remain insufficient: they either rely on external reference data or fail to be bo...

Scaling DeFi with ZK Rollups: Design, Deployment, and Evaluation of a Real-Time Proof-Of-Concept

Ethereum's scalability limitations pose significant challenges for the adoption of decentralized applications dApps. Zero-Knowledge Rollups ZK Rollups present a promising solution, bundling transactions off-chain and submitting validity proofs on-chain to enhance throughput and efficiency. In thi...

Con Instruction: Universal Jailbreaking of Multimodal Large Language Models Via Non-Textual Modalities

Existing attacks against multimodal language models MLLMs primarily communicate instructions through text accompanied by adversarial images. In contrast, we exploit the capabilities of MLLMs to interpret non-textual instructions, specifically, adversarial images or audio generated by our novel...

Dpmm: Differentially Private Marginal Models, a Library for Synthetic Tabular Data Generation

We propose dpmm, an open-source library for synthetic data generation with Differentially Private DP guarantees. It includes three popular marginal models -- PrivBayes, MST, and AIM -- that achieve superior utility and offer richer functionality compared to alternative implementations...

Adversarial Threat Vectors and Risk Mitigation for Retrieval-Augmented Generation Systems

Retrieval-Augmented Generation RAG systems, which integrate Large Language Models LLMs with external knowledge sources, are vulnerable to a range of adversarial attack vectors. This paper examines the importance of RAG systems through recent industry adoption trends and identifies the prominent...

Data Flows in You: Benchmarking and Improving Static Data-Flow Analysis on Binary Executables

Data-flow analysis is a critical component of security research. Theoretically, accurate data-flow analysis in binary executables is an undecidable problem, due to complexities of binary code. Practically, many binary analysis engines offer some data-flow analysis capability, but we lack...

Eve File Disclosure / Code Execution

Eve versions prior to 0.7.5 blind remote code execution proof of concept that retrieves files...

Chainless Apps: a Modular Framework for Building Apps with Web2 Capability and Web3 Trust

Modern blockchain applications are often constrained by a trade-off between user experience and trust. Chainless Apps present a new paradigm of application architecture that separates execution, trust, bridging, and settlement into distinct compostable layers. This enables app-specific sequencing...

Test-Time Immunization: a Universal Defense Framework against Jailbreaks for (Multimodal) Large Language Models

While multimodal large language models LLMs have attracted widespread attention due to their exceptional capabilities, they remain vulnerable to jailbreak attacks. Various defense methods are proposed to defend against jailbreak attacks, however, they are often tailored to specific types of...

BugWhisperer: Fine-Tuning LLMs for SoC Hardware Vulnerability Detection

The current landscape of system-on-chips SoCs security verification faces challenges due to manual, labor-intensive, and inflexible methodologies. These issues limit the scalability and effectiveness of security protocols, making bug detection at the Register-Transfer Level RTL difficult. This...

Evaluating AI Cyber Capabilities with Crowdsourced Elicitation

As AI systems become increasingly capable, understanding their offensive cyber potential is critical for informed governance and responsible deployment. However, it's hard to accurately bound their capabilities, and some prior evaluations dramatically underestimated them. The art of extracting...

AdInject: Real-World Black-Box Attacks on Web Agents Via Advertising Delivery

Vision-Language Model VLM based Web Agents represent a significant step towards automating complex tasks by simulating human-like interaction with websites. However, their deployment in uncontrolled web environments introduces significant security vulnerabilities. Existing research on adversarial...

A Joint Reconstruction-Triplet Loss Autoencoder Approach Towards Unseen Attack Detection in IoV Networks

Internet of Vehicles IoV systems, while offering significant advancements in transportation efficiency and safety, introduce substantial security vulnerabilities due to their highly interconnected nature. These dynamic systems produce massive amounts of data between vehicles, infrastructure, and...

Uncovering Black-Hat SEO Based Fake E-Commerce Scam Groups from Their Redirectors and Websites

While law enforcements agencies and cybercrime researchers are working hard, fake E-commerce scam is still a big threat to Internet users. One of the major techniques to victimize users is luring them by black-hat search-engine-optimization SEO; making search engines display their lure pages as i...

DP-RTFL: Differentially Private Resilient Temporal Federated Learning for Trustworthy AI in Regulated Industries

Federated Learning FL has emerged as a critical paradigm for enabling privacy-preserving machine learning, particularly in regulated sectors such as finance and healthcare. However, standard FL strategies often encounter significant operational challenges related to fault tolerance, system...

CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models

Retrieval-Augmented Generation RAG enhances large language models LLMs by incorporating external knowledge, but its openness introduces vulnerabilities that can be exploited by poisoning attacks. Existing poisoning methods for RAG systems have limitations, such as poor generalization and lack of...

TeleSparse: Practical Privacy-Preserving Verification of Deep Neural Networks

Verification of the integrity of deep learning inference is crucial for understanding whether a model is being applied correctly. However, such verification typically requires access to model weights and potentially sensitive or private training data. So-called Zero-knowledge Succinct...

Usability of Token-Based and Remote Electronic Signatures: a User Experience Study

As electronic signatures e-signatures become increasingly integral to secure digital transactions, understanding their usability and security perception from an end-user perspective has become crucial. This study empirically evaluates and compares two major e-signature systems -- token-based and...

Mal-D2GAN: Double-Detector Based GAN for Malware Generation

Machine learning ML has been developed to detect malware in recent years. Most researchers focused their efforts on improving the detection performance but ignored the robustness of the ML models. In addition, many machine learning algorithms are very vulnerable to intentional attacks. To solve...

Invisible Tokens, Visible Bills: the Urgent Need to Audit Hidden Operations in Opaque LLM Services

Whitepaper called Invisible Tokens, Visible Bills: The Urgent Need To Audit Hidden Operations In Opaque LLM Services...

Finetuning-Activated Backdoors in LLMs

Finetuning openly accessible Large Language Models LLMs has become standard practice for achieving task-specific performance improvements. Until now, finetuning has been regarded as a controlled and secure process in which training on benign datasets led to predictable behaviors. In this paper, w...

Harry Potter Is Still Here! Probing Knowledge Leakage in Targeted Unlearned Large Language Models Via Automated Adversarial Prompting

This work presents LURK Latent UnleaRned Knowledge, a novel framework that probes for hidden retained knowledge in unlearned LLMs through adversarial suffix prompting. LURK automatically generates adversarial prompt suffixes designed to elicit residual knowledge about the Harry Potter domain, a...

Unlearning Isn'T Deletion: Investigating Reversibility of Machine Unlearning in LLMs

Unlearning in large language models LLMs is intended to remove the influence of specific data, yet current evaluations rely heavily on token-level metrics such as accuracy and perplexity. We show that these metrics can be misleading: models often appear to forget, but their original behavior can ...

TP-Link Archer AX50 Buffer Overflow

The TP-Link Archer AX50 router is vulnerable to a stack-based buffer overflow on its firmware version 1.0.14 Build 20240108 rel.426554555, leading to remote code execution both in the LAN and in the WAN side. This vulnerability is the same as CVE-2020-10881, found by the Flashback team and largel...

WordPress Madara 2.2.2 Local File Inclusion

WordPress Madara theme versions 2.2.2 and below suffer from a local file inclusion vulnerability...

Alignment under Pressure: the Case for Informed Adversaries When Evaluating LLM Defenses

Large language models LLMs are rapidly deployed in real-world applications ranging from chatbots to agentic systems. Alignment is one of the main approaches used to defend against attacks such as prompt injection and jailbreaks. Recent defenses report near-zero Attack Success Rates ASR even again...

Mitigating Cyber Risk in the Age of Open-Weight LLMs: Policy Gaps and Technical Realities

Open-weight general-purpose AI GPAI models offer significant benefits but also introduce substantial cybersecurity risks, as demonstrated by the offensive capabilities of models like DeepSeek-R1 in evaluations such as MITRE's OCCULT. These publicly available models empower a wider range of actors...

Quantum-Resilient Blockchain for Secure Transactions in UAV-Assisted Smart Agriculture Networks

The integration of unmanned aerial vehicles UAVs into smart agriculture has enabled real-time monitoring, data collection, and automated farming operations. However, the high mobility, decentralized nature, and low-power communication of UAVs pose significant security challenges, particularly in...

Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems through Benign Queries

Retrieval-Augmented Generation RAG systems enhance large language models LLMs by incorporating external knowledge bases, but they are vulnerable to privacy risks from data extraction attacks. Existing extraction methods typically rely on malicious inputs such as prompt injection or jailbreaking,...

Outsourcing SAT-Based Verification Computations in Network Security

The emergence of cloud computing gives huge impact on large computations. Cloud computing platforms offer servers with large computation power to be available for customers. These servers can be used efficiently to solve problems that are complex by nature, for example, satisfiability SAT problem...

D4+: Emergent Adversarial Driving Maneuvers with Approximate Functional Optimization

Intelligent mechanisms implemented in autonomous vehicles, such as proactive driving assist and collision alerts, reduce traffic accidents. However, verifying their correct functionality is difficult due to complex interactions with the environment. This problem is exacerbated in adversarial...

Is Your Prompt Safe? Investigating Prompt Injection Attacks against Open-Source LLMs

Whitepaper called Is Your Prompt Safe? Investigating Prompt Injection Attacks Against Open-Source LLMs...

Robust and Efficient AI-Based Attack Recovery in Autonomous Drones

We introduce an autonomous attack recovery architecture to add common sense reasoning to plan a recovery action after an attack is detected. We outline use-cases of our architecture using drones, and then discuss how to implement this architecture efficiently and securely in edge devices...

Recommender Systems for Democracy: toward Adversarial Robustness in Voting Advice Applications

Voting advice applications VAAs help millions of voters understand which political parties or candidates best align with their views. This paper explores the potential risks these applications pose to the democratic process when targeted by adversarial entities. In particular, we expose 11...

DynaNoise: Dynamic Probabilistic Noise Injection for Defending against Membership Inference Attacks

Membership Inference Attacks MIAs pose a significant risk to the privacy of training datasets by exploiting subtle differences in model outputs to determine whether a particular data sample was used during training. These attacks can compromise sensitive information, especially in domains such as...

MorphMark: Flexible Adaptive Watermarking for Large Language Models

Watermarking by altering token sampling probabilities based on red-green list is a promising method for tracing the origin of text generated by large language models LLMs. However, existing watermark methods often struggle with a fundamental dilemma: improving watermark effectiveness the...

ACU: Analytic Continual Unlearning for Efficient and Exact Forgetting with Privacy Preservation

The development of artificial intelligence demands that models incrementally update knowledge by Continual Learning CL to adapt to open-world environments. To meet privacy and security requirements, Continual Unlearning CU emerges as an important problem, aiming to sequentially forget particular...

Security Practices in AI Development

What makes safety claims about general purpose AI systems such as large language models trustworthy? We show that rather than the capabilities of security tools such as alignment and red teaming procedures, it is security practices based on these tools that contributed to reconfiguring the image ...

Efficient Implementations of Residue Generators Mod 2n + 1 Providing Diminished-1 Representation

The moduli of the form 2n + 1 belong to a class of low-cost odd moduli, which have been frequently selected to form the basis of various residue number systems RNS. The most efficient computations modulo mod 2n + 1 are performed using the so-called diminished-1 D1 representation. Therefore, it is...

Nosy Layers, Noisy Fixes: Tackling DRAs in Federated Learning Systems Using Explainable AI

Federated Learning FL has emerged as a powerful paradigm for collaborative model training while keeping client data decentralized and private. However, it is vulnerable to Data Reconstruction Attacks DRA such as "LoKI" and "Robbing the Fed", where malicious models sent from the server to the clie...

AutoRAN: Weak-To-Strong Jailbreaking of Large Reasoning Models

This paper presents AutoRAN, the first automated, weak-to-strong jailbreak attack framework targeting large reasoning models LRMs. At its core, AutoRAN leverages a weak, less-aligned reasoning model to simulate the target model's high-level reasoning structures, generates narrative prompts, and...

Side Channel Analysis in Homomorphic Encryption

Homomorphic encryption provides many opportunities for privacy-aware processing, including with methods related to machine learning. Many of our existing cryptographic methods have been shown in the past to be susceptible to side channel attacks. With these, the implementation of the cryptographi...