相关论文: IssueGuard: Real-Time Secret Leak Prevention Tool …
This paper describes MessageGuard, a browser-based platform for research into usable content-based encryption. MessageGuard is designed to enable collaboration between security and usability researchers on long-standing research questions…
The potential misuse of ChatGPT and other Large Language Models (LLMs) has raised concerns regarding the dissemination of false information, plagiarism, academic dishonesty, and fraudulent activities. Consequently, distinguishing between…
Static analysis has established itself as a weapon of choice for detecting security vulnerabilities. Taint analysis in particular is a very general and powerful technique, where security policies are expressed in terms of forbidden flows,…
The nature of personalized text-to-image models poses a unique safety challenge that generic context-blind methods are ill-equipped to handle. Such global filters create a dilemma: to prevent misuse, they are forced to damage the model's…
We present CureSpec, the first model-checking based framework for automatic repair of programs with respect to information leaks in the presence of side-channels and speculative execution. CureSpec is based on formal models of attacker…
Modern AI agents execute real-world side effects through tool calls such as file operations, shell commands, HTTP requests, and database queries. A single unsafe action, including accidental deletion, credential exposure, or data…
Backdoor attacks have become a major security threat for deploying machine learning models in security-critical applications. Existing research endeavors have proposed many defenses against backdoor attacks. Despite demonstrating certain…
Database research and development rely heavily on realistic user workloads for benchmarking, instance optimization, migration testing, and database tuning. However, acquiring real-world SQL queries is notoriously challenging due to strict…
Toxicity detection mitigates the dissemination of toxic content (e.g., hateful comments, posts, and messages within online social actions) to safeguard a healthy online social environment. However, malicious users persistently develop…
In this research, we use user defined labels from three internet text sources (Reddit, Stackexchange, Arxiv) to train 21 different machine learning models for the topic classification task of detecting cybersecurity discussions in natural…
Language generation models have been an increasingly powerful enabler for many applications. Many such models offer free or affordable API access, which makes them potentially vulnerable to model extraction attacks through distillation. To…
Authorship verification (AV) aims to identify whether a pair of texts has the same author. We address the challenge of evaluating AV models' robustness against topic shifts. The conventional evaluation assumes minimal topic overlap between…
Modern automated accessibility testing tools for mobile applications have significantly improved the detection of interface violations, yet their impact on remediation remains limited. A key reason is that existing tools typically produce…
This paper introduces LadyBug, a GitHub bot that automatically localizes bugs for Android apps by combining UI interaction information with text retrieval. LadyBug connects to an Android app's GitHub repository, and is triggered when a bug…
Real-time systems have recently been shown to be vulnerable to timing inference attacks, mainly due to their predictable behavioral patterns. Existing solutions such as schedule randomization lack the ability to protect against such…
While extremely valuable to achieve advanced functions, mobile phone sensors can be abused by attackers to implement malicious activities in Android apps, as experimentally demonstrated by many state-of-the-art studies. There is hence a…
Organizations that scan documents for sensitive information face a practical problem. Cloud services require data to be sent to external infrastructure, while rule-based tools often miss threats that depend on context. This study presents…
The rapid proliferation of Artificial Intelligence Generated Content has precipitated a crisis of trust and urgent regulatory demands. However, existing identification tools suffer from fragmentation and a lack of support for visible…
Intel SGX (Software Guard Extension) is a promising TEE (trusted execution environment) technique that can protect programs running in user space from being maliciously accessed by the host operating system. Although it provides hardware…
The demand for quick and reliable DevOps operations pushed distributors of repository platforms to implement workflows. Workflows allow automating code management operations directly on the repository hosting the software. However, this…