相关论文: Context-Auditor: Context-sensitive Content Injecti…

Learned classifiers deployed in agentic pipelines face a fundamental reliability problem: predictions are probabilistic inferences, not verified conclusions, and acting on them without grounding in observable evidence leads to compounding…

With the development of blockchain technology, the detection of smart contract vulnerabilities is increasingly emphasized. However, when detecting vulnerabilities in inter-contract interactions (i.e., cross-contract vulnerabilities) using…

Detection and mitigation of critical web vulnerabilities and attacks like cross-site scripting (XSS), and cross-site request forgery (CSRF) have been a great concern in the field of web security. Such web attacks are evolving and becoming…

A popular class of defenses against prompt injection attacks on large language models (LLMs) relies on fine-tuning to separate instructions and data, so that the LLM does not follow instructions that might be present with data. We evaluate…

Contextual bandits are widely used in industrial personalization systems. These online learning frameworks learn a treatment assignment policy in the presence of treatment effects that vary with the observed contextual features of the…

Large language model (LLM) agents increasingly rely on external tools and retrieval systems to autonomously complete complex tasks. However, this design exposes agents to indirect prompt injection (IPI), where attacker-controlled context…

With the widespread use of deep learning system in many applications, the adversary has strong incentive to explore vulnerabilities of deep neural networks and manipulate them. Backdoor attacks against deep neural networks have been…

Due to the high cost of training, large model (LM) practitioners commonly use pretrained models downloaded from untrusted sources, which could lead to owning compromised models. In-context learning is the ability of LMs to perform multiple…

Neural contextual biasing effectively improves automatic speech recognition (ASR) for crucial phrases within a speaker's context, particularly those that are infrequent in the training data. This work proposes contextual text injection…

Check Point researchers revealed a new attack vector which threatens millions of users worldwide - attack by subtitles. By crafting malicious subtitle files, which are then downloaded by a victim's media player, attackers can take complete…

Several years of research have shown that machine-learning systems are vulnerable to adversarial examples, both in theory and in practice. Until now, such attacks have primarily targeted visual models, exploiting the gap between human and…

Conversational speech synthesis (CSS) incorporates historical dialogue as supplementary information with the aim of generating speech that has dialogue-appropriate prosody. While previous methods have already delved into enhancing context…

PowerShell is a command-line shell, supporting a scripting language. It is widely used in organizations for configuration management and task automation but is also increasingly used by cybercriminals for launching cyberattacks against…

Acoustic side-channel attacks on keyboards can bypass security measures in many systems that use keyboards as one of the input devices. These attacks aim to reveal users' sensitive information by targeting the sounds made by their keyboards…

Deep learning models have achieved great success in many fields, yet they are vulnerable to adversarial examples. This paper follows a causal perspective to look into the adversarial vulnerability and proposes Causal Intervention by…

The rapid growth of natural language processing (NLP) and pre-trained language models have enabled accurate text classification in a variety of settings. However, text classification models are susceptible to backdoor attacks, where an…

Browser agents enable autonomous web interaction but face critical reliability and security challenges in production. This paper presents findings from building and operating a production browser agent. The analysis examines where current…

Prompt injection attacks can compromise the security and stability of critical systems, from infrastructure to large web applications. This work curates and augments a prompt injection dataset based on the HackAPrompt Playground Submissions…

Autonomous UI agents powered by AI have tremendous potential to boost human productivity by automating routine tasks such as filing taxes and paying bills. However, a major challenge in unlocking their full potential is security, which is…

Users in various web and mobile applications are vulnerable to attribute inference attacks, in which an attacker leverages a machine learning classifier to infer a target user's private attributes (e.g., location, sexual orientation,…