Ruse

Ruse Mobile camera-based application that attempts to alter photos to preserve their utility to humans while making them unusable for facial recognition systems. Installation 1 Easy Method: Wait and download app from appropriate app store. 2 Download and run ios app via XCode see Development setu...

OpenAttack

Documentation • Features & Uses • Usage Examples • Attack Models • Toolkit Design OpenAttack is an open-source Python-based textual adversarial attack toolkit, which handles the whole process of textual adversarial attacking, including preprocessing text, accessing the victim model, generating...

Anti-Tamper-Perturbation

🛡️ Anti-Tamper Protection for Unauthorized Individual Image Generation Official Code for ICCV 2025 Paper 📄 Anti-Tamper Protection for Unauthorized Individual Image Generation ⚙️ Environment Setup conda create --name ATP --file requirements.txt conda activate ATP 🔐 Launch ATP 1. Download Dataset...

Perturbed-Embedding-Vectors

Jailbreaking Open-Weight LLMs via Random Embedding Perturbations: runs, labels and code Code and data for the paper Jailbreaking Open-Weight LLMs via Random Embedding Perturbations. Authored by the amazing people of the Theoretical Computer Science Department of University of California, Santa...

Region-Level Black-Box Defense against Stealthy Embedding-Space Backdoors in CLIP

Contrastive Language--Image Pretraining CLIP has emerged as a dominant vision backbone due to its strong transferability and zero-shot capabilities. However, recent studies reveal a critical vulnerability: embedding-space backdoor attacks. By poisoning only a tiny fraction of image--text pairs,...

A Novel Steganography Scheme Using Quantum Hilbert Transform

The main goal of steganography is to transmit hidden messages in legitimate-looking communication messages. Phase-domain information hiding, however, has not been fully explored for quantum systems. This work introduces a finite-dimensional Quantum Hilbert Transform QHT as a unitary phase operato...

CAPTCHAs in the Agentic Era: Solvers That Learn from Every Encounter

Vision-language models VLMs can solve visual CAPTCHAs without task-specific training, but the agents built on them approach every challenge from scratch. For such an agent, the hundredth instance of a familiar puzzle costs as much time and compute as the first. Specialized detectors invert the...

Vulnerable Code Search: Transferable Attack for Code Language Models

Reliable code retrieval is crucial for developer productivity and effective code reuse. However, current neural code language models CLMs powering search tools are susceptible to adversarial attacks targeting non-functional textual elements. In this paper, we introduce a programming...

On Understanding, Identifying, and Mitigating Vulnerabilities in Agentic Large Language Models

Large Language Models LLMs have undergone a shift from stateless conversational interfaces to autonomous agents capable of multi-step planning, tool invocation, code execution, and maintaining persistent memory. When these agents operate with real-world privileges---calling APIs, modifying files,...

Never Stop Speaking: A Denial-Of-Service Attack on End-To-End Speech Language Models

Many studies have shown that specially crafted inputs can induce large language models LLMs to generate excessively long outputs, resulting in significant computational overhead and resource consumption. While most existing denial-of-service DoS attacks target text-only LLMs, end-to-end E2E speec...

Beyond Resilience: Antifragility in Critical Infrastructure Cybersecurity

Critical infrastructure cybersecurity increasingly requires frameworks that move beyond recovery toward bounded improvement under disruption, yet empirically grounded theories for operational technology remain limited. This paper develops a Theory of Antifragility AFT for critical infrastructure ...

Geometric Configurations of Perturbed Jailbreak Prompts

Perturbation techniques that turn unsuccessful jailbreak prompts into successful ones are continuously evolving, constituting a major security threat to LLM safety. In this paper, we investigate the internal representations of such string-level perturbed jailbreak inputs in the small weight model...

ReDoS in giskard's transformation.py (GHSL-2024-324)

ReDoS in Giskard text perturbation detectorA Remote Code Execution ReDoS vulnerability was discovered in Giskard component by the GitHub Security Lab team. When processing datasets with specific text patterns with Giskard detectors, this vulnerability could trigger exponential regex evaluation...

PYSEC-2026-1403 ReDoS in giskard's transformation.py (GHSL-2024-324)

ReDoS in Giskard text perturbation detector A Remote Code Execution ReDoS vulnerability was discovered in Giskard component by the GitHub Security Lab team. When processing datasets with specific text patterns with Giskard detectors, this vulnerability could trigger exponential regex evaluation...

Detecting Adversarial Evasion Attacks against Autoencoder-Based Network Intrusion Detection Systems

Evasion attacks deliberately manipulate input to an ML-based system to produce an incorrect prediction while the manipulated input still appears benign. The PANDA framework has demonstrated that adversarial examples developed for the vision domain can be transferred to the network domain by...

FlipGuard: Defending Large Language Models against Quantization-Conditioned Backdoor Attacks

Model quantization is essential for the efficient deployment of Large Language Models LLMs, but introduces a critical vulnerability: Quantization-Conditioned Backdoor QCB attacks. In these attacks, malicious behaviors remain dormant in full-precision models and activate only after specific...

DarkLLM: Learning Language-Driven Adversarial Attacks with Large Language Models

While vision and multimodal foundation models underpin critical tasks from perception to complex reasoning, they remain highly vulnerable to adversarial attacks. However, traditional adversarial attacks are typically limited to single, predefined objectives, tightly coupling each attack to a...

ID-Eraser: Proactive Defense against Face Swapping Via Identity Perturbation

Deepfake technologies have rapidly advanced with modern generative AI, and face swapping in particular poses serious threats to privacy and digital security. Existing proactive defenses mostly rely on pixel-level perturbations, which are ineffective against contemporary swapping models that extra...

Jailbreaking the Matrix: Nullspace Steering for Controlled Model Subversion

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 HMNS, a circuit-level intervention that i identifies attentio...

RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance

Deep neural networks are highly susceptible to backdoor attacks, yet most defense methods to date rely on balanced data, overlooking the pervasive class imbalance in real-world scenarios that can amplify backdoor threats. This paper presents the first in-depth investigation of how the dataset...