Phare: a Safety Probe for Large Language Models

Ensuring the safety of large language models LLMs is critical for responsible deployment, yet existing evaluations often prioritize performance over identifying failure modes. We introduce Phare, a multilingual diagnostic framework to probe and evaluate LLM behavior across three critical...

CVE-2024-44006

Missing Authorization vulnerability in Amir Helzer WooCommerce Multilingual & Multicurrency woocommerce-multilingual.This issue affects WooCommerce Multilingual & Multicurrency: from n/a through = 5.3.6...

CVE-2022-3141

The Translate Multilingual sites WordPress plugin before 2.3.3 is vulnerable to an authenticated SQL injection. By adding a new language via the settings page containing specific special characters, the backticks in the SQL query can be surpassed and a time-based blind payload can be injected...

CVE-2020-10568

The sitepress-multilingual-cms WPML plugin before 4.3.7-b.2 for WordPress has CSRF due to a loose comparison. This leads to remote code execution in includes/class-wp-installer.php via a series of requests that leverage unintended comparisons of integers to strings...

CVE-2019-1010290

Babel: Multilingual site Babel All is affected by: Open Redirection. The impact is: Redirection to any URL, which is supplied to redirect.php in a "newurl" parameter. The component is: redirect.php. The attack vector is: The victim must open a link created by an attacker. Attacker may use any...

CVE-2015-9416

The sitepress-multilingual-cms WPML plugin 2.9.3 to 3.2.6 for WordPress has XSS via the Accept-Language HTTP header...

MAPS: a Multilingual Benchmark for Global Agent Performance and Security

Agentic AI systems, which build on Large Language Models LLMs and interact with tools and memory, have rapidly advanced in capability and scope. Yet, since LLMs have been shown to struggle in multilingual settings, typically resulting in lower performance and reduced safety, agentic systems risk...

BadLingual: a Novel Lingual-Backdoor Attack against Large Language Models

In this paper, we present a new form of backdoor attack against Large Language Models LLMs: lingual-backdoor attacks. The key novelty of lingual-backdoor attacks is that the language itself serves as the trigger to hijack the infected LLMs to generate inflammatory speech. They enable the precise...

WordPress plugin WPML 跨站脚本漏洞

WordPress and WordPress plugin are both products of the WordPress Foundation.WordPress is a set of blogging platform developed using the PHP language. The platform supports setting up personal blog sites on servers with PHP and MySQL.WordPress plugin is an application plugin. A cross-site scripti...

Darcula Phishing Kit Uses AI to Evade Detection, Experts Warn

Darcula phishing platform adds AI to create multilingual scam pages easily. Netcraft warns of rising risks from Darcula-Suite…...

The vulnerability of the NVIDIA Riva multilingual speech processing and translation tools is related to access control deficiencies, which allow attackers to trigger service failures.

The vulnerability of the NVIDIA Riva multilingual speech processing and translation tools is related to deficiencies in access control. Exploiting this vulnerability could allow a malicious actor to cause service failures...

Concept Enhancement Engineering: a Lightweight and Efficient Robust Defense against Jailbreak Attacks in Embodied AI

Embodied Intelligence EI systems integrated with large language models LLMs face significant security risks, particularly from jailbreak attacks that manipulate models into generating harmful outputs or executing unsafe physical actions. Traditional defense strategies, such as input filtering and...

CVE-2025-26888 WordPress WooCommerce Multilingual & Multicurrency plugin <= 5.3.8 - Broken Access Control vulnerability

Missing Authorization vulnerability in Amir Helzer WooCommerce Multilingual & Multicurrency woocommerce-multilingual allows Exploiting Incorrectly Configured Access Control Security Levels.This issue affects WooCommerce Multilingual & Multicurrency: from n/a through = 5.3.8...

WordPress WooCommerce Multilingual & Multicurrency plugin <= 5.3.8 - Broken Access Control vulnerability

Broken Access Control vulnerability discovered by Rafie Muhammad Patchstack in WordPress Plugin WooCommerce Multilingual & Multicurrency versions = 5.3.8...

WordPress plugin WooCommerce Multilingual & Multicurrency 安全漏洞

WordPress and WordPress plugin are both products of the WordPress Foundation.WordPress is a blogging platform developed using the PHP language. The platform supports setting up personal blog sites on servers with PHP and MySQL.WordPress plugin is an application plugin. A security vulnerability...

F5 NGINX Unit 安全漏洞

F5 NGINX Unit is a multilingual dynamic application server from F5 Corporation. The product supports a RESTful JSON API to run applications built using multiple languages and frameworks. A security vulnerability exists in F5 NGINX Unit prior to version 1.34.2, which stems from an infinite loop th...

DRUPAL-CONTRIB-2025-014

Open Social is a Drupal distribution for online communities, which ships with a default optional module social\language to make your platform multilingual. Some site administration configuration does not correctly check access when trying to translate allowing unauthorised people to translate the...

Open Social - Moderately critical - Access bypass - SA-CONTRIB-2025-014

Open Social is a Drupal distribution for online communities, which ships with a default optional module sociallanguage to make your platform multilingual. Some site administration configuration does not correctly check access when trying to translate allowing unauthorised people to translate thes...

CVE-2024-32602

Improper Neutralization of Special Elements used in an SQL Command 'SQL Injection' vulnerability in OnTheGoSystems WooCommerce Multilingual & Multicurrency.This issue affects WooCommerce Multilingual & Multicurrency: from n/a through 5.3.3.1...

PMD 安全漏洞

PMD is an extensible multilingual static source code analyzer. A security vulnerability exists in PMD that stems from a possible disclosure of the password for issuing signing keys...