About the security content of visionOS 2.6

About the security content of visionOS 2.6 This document describes the security content of visionOS 2.6. About Apple security updates For our customers' protection, Apple doesn't disclose, discuss, or confirm security issues until an investigation has occurred and patches or releases are availabl...

In-Context Learning of Vision Language Models for Detection of Physical and Digital Attacks against Face Recognition Systems

Recent advances in biometric systems have significantly improved the detection and prevention of fraudulent activities. However, as detection methods improve, attack techniques become increasingly sophisticated. Attacks on face recognition systems can be broadly divided into physical and digital...

CVE-2025-40985

SQL injection vulnerability in SCATI Vision Web of SCATI Labs from version 4.8 to 7.2. This vulnerability allows an attacker to exfiltrate some data from the database via the ‘login’ parameter in the endpoint ‘/scatevisionweb/index.php/loginForm’...

Breaking the Illusion of Security Via Interpretation: Interpretable Vision Transformer Systems under Attack

Vision transformer ViT models, when coupled with interpretation models, are regarded as secure and challenging to deceive, making them well-suited for security-critical domains such as medical applications, autonomous vehicles, drones, and robotics. However, successful attacks on these systems ca...

CVE-2025-53644 OpenCV contains a use after free buffer write due to an uninitialized pointer

OpenCV is an Open Source Computer Vision Library. Versions 4.10.0 and 4.11.0 have an uninitialized pointer variable on stack that may lead to arbitrary heap buffer write when reading crafted JPEG images. Version 4.12.0 fixes the vulnerability...

CVE-2025-40985

SQL injection vulnerability in SCATI Vision Web of SCATI Labs from version 4.8 to 7.2. This vulnerability allows an attacker to exfiltrate some data from the database via the ‘login’ parameter in the endpoint ‘/scatevisionweb/index.php/loginForm’...

CVE-2025-40985 SQL Injection in SCATI Vision Web

SQL injection vulnerability in SCATI Vision Web of SCATI Labs from version 4.8 to 7.2. This vulnerability allows an attacker to exfiltrate some data from the database via the ‘login’ parameter in the endpoint ‘/scatevisionweb/index.php/loginForm’...

CVE-2025-40985 SQL Injection in SCATI Vision Web

SQL injection vulnerability in SCATI Vision Web of SCATI Labs from version 4.8 to 7.2. This vulnerability allows an attacker to exfiltrate some data from the database via the ‘login’ parameter in the endpoint ‘/scatevisionweb/index.php/loginForm’...

CVE-2025-40985

Summary of CVE-2025-40985 (SCATI Vision Web) : A SQL injection vulnerability affects SCATI Vision Web versions 4.8 through 7.2. The flaw enables an attacker to exfiltrate data from the database via the loginForm endpoint (/scatevision_web/index.php/loginForm). The provided documents consistently ...

SCATI Vision Web SQL注入漏洞

SCATI Vision Web is a browser component from SCATI Spain. A SQL injection vulnerability exists in SCATI Vision Web versions 4.8 through 7.2, which originates from a SQL injection and could lead to the disclosure of database information...

PT-2025-29714 · Scati · Scati Vision Web

Name of the Vulnerable Software and Affected Versions: SCATI Vision Web versions 4.8 through 7.2 Description: A SQL injection issue exists in SCATI Vision Web. This vulnerability allows an attacker to exfiltrate data from the database via the login parameter in the /scatevision...

Bridging the Gap in Vision Language Models in Identifying Unsafe Concepts across Modalities

Whitepaper called Bridging The Gap In Vision Language Models In Identifying Unsafe Concepts Across Modalities...

CVE-2025-27058 Buffer Copy Without Checking Size of Input in Computer Vision

Memory corruption while processing packet data with exceedingly large packet...

CVE-2025-27058 Buffer Copy Without Checking Size of Input in Computer Vision

Memory corruption while processing packet data with exceedingly large packet...

BackFed: an Efficient and Standardized Benchmark Suite for Backdoor Attacks in Federated Learning

Federated Learning FL systems are vulnerable to backdoor attacks, where adversaries train their local models on poisoned data and submit poisoned model updates to compromise the global model. Despite numerous proposed attacks and defenses, divergent experimental settings, implementation errors, a...

UniAud: a Unified Auditing Framework for High Auditing Power and Utility with One Training Run

Differentially private DP optimization has been widely adopted as a standard approach to provide rigorous privacy guarantees for training datasets. DP auditing verifies whether a model trained with DP optimization satisfies its claimed privacy level by estimating empirical privacy lower bounds...

The Age of Sensorial Zero Trust: Why We Can No Longer Trust Our Senses

In a world where deepfakes and cloned voices are emerging as sophisticated attack vectors, organizations require a new security mindset: Sensorial Zero Trust. This article presents a scientific analysis of the need to systematically doubt information perceived through the senses, establishing...

Boosting Generative Adversarial Transferability with Self-Supervised Vision Transformer Features

The ability of deep neural networks DNNs come from extracting and interpreting features from the data provided. By exploiting intermediate features in DNNs instead of relying on hard labels, we craft adversarial perturbation that generalize more effectively, boosting black-box transferability...

On the Feasibility of Poisoning Text-To-Image AI Models Via Adversarial Mislabeling

Today's text-to-image generative models are trained on millions of images sourced from the Internet, each paired with a detailed caption produced by Vision-Language Models VLMs. This part of the training pipeline is critical for supplying the models with large volumes of high-quality image-captio...

E-FreeM2: Efficient Training-Free Multi-Scale and Cross-Modal News Verification Via MLLMs

The rapid spread of misinformation in mobile and wireless networks presents critical security challenges. This study introduces a training-free, retrieval-based multimodal fact verification system that leverages pretrained vision-language models and large language models for credibility assessmen...