密码学与安全
Embedding-as-a-Service (EaaS) has become an important semantic infrastructure for natural language and multimedia applications, but it is highly vulnerable to model stealing and copyright infringement. Existing EaaS watermarking methods…
IoT botnet detection has advanced, yet most published systems are validated on a single dataset and rarely generalise across environments. Heterogeneous feature spaces make multi-dataset training practically impossible without discarding…
Large Language Models (LLMs) face prominent security risks from jailbreaking, a practice that manipulates models to bypass built-in security constraints and generate unethical or unsafe content. Among various jailbreak techniques,…
Over the years, many techniques have been introduced to protect integrated circuits (ICs) from hardware security threats that emerged in the globalized IC manufacturing supply chain, such as overproduction and piracy. However, most of these…
Existing methods for detection rule generation are tightly coupled to specific input-output combinations, requiring dedicated pipelines for each. We formalize this problem as a unified mapping f:C*L->R and characterize optimal rules through…
Deep neural networks remain highly vulnerable to adversarial perturbations, limiting their reliability in security- and safety-critical applications. To address this challenge, we introduce QShield, a modular hybrid quantum-classical neural…
Watermarking provides a critical safeguard for large language model (LLM) services by facilitating the detection of LLM-generated text. Correspondingly, stealing watermark algorithms (SWAs) derive watermark information from watermarked…
In recent years, many countries have started enacting laws to safeguard privacy of personal data of their citizens collected and maintained by various enterprises through websites, mobile apps, and other means. It is imperative that the…
Retrieval-Augmented Generation (RAG) systems augment large language models with external knowledge, yet introduce a critical security vulnerability: RAG Knowledge Base Leakage, wherein adversarial prompts can induce the model to divulge…
The Self-Sovereign Identity (SSI) paradigm is instrumental for decentralised identity management, allowing an entity to create, manage, and present their digital credentials without relying on centralised authorities. Credential selective…
Downfall is a side-channel attack that leaks values in vector registers from a process to another on the same CPU core. This attack enables an attacker to achieve serious outcomes (e.g., stealing AES keys), and there is no fundamental…
The Model Context Protocol (MCP) is a new and emerging technology that extends the functionality of large language models, improving workflows but also exposing users to a new attack surface. Several studies have highlighted related…
Deepfake content on social networks is increasingly produced through multiple \emph{sequential} edits to biometric data such as facial imagery. Consequently, the final appearance of an image often reflects a latent chain of operations…
Recent advances in research on Attribute-based Access Control (ABAC) has led to the development of several ingenious methods for representing and enforcing organizational security policies. However, so far little effort has been spent…
We give a public key encryption scheme with plausible quasi-exponential security based on the conjectured intractability of two constraint satisfaction problems (CSPs), both of which are instantiated with a corruption rate of $1 - o(1)$.…
As artificial intelligence (AI) systems grow more powerful, autonomous, and embedded in critical infrastructure, their identification and traceability become foundational to regulatory oversight and sustainable digital governance. In…
Electronic cash (e-cash) is a digital alternative to physical currency that allows anonymous transactions between users and merchants. Typically, coins in an e-cash scheme are only dispensed through a central bank. A drawback of this…
Large language models remain vulnerable to jailbreak attacks -- inputs designed to bypass safety mechanisms and elicit harmful responses -- despite advances in alignment and instruction tuning. We propose Head-Masked Nullspace Steering…
We provide an approach that closely estimates an organization's cyber resources directly from vulnerability timestamps, using a non-stationary queueing framework. Traditional attack-surface metrics operate on static snapshots, ignoring the…
Toxicity and harassment are widespread in the video-gaming context. Especially in competitive online multiplayer scenarios, gamers oftentimes send harmful messages to other players (teammates or opponents) whose consequences span from mild…