Winter Soldier: Backdooring Language Models at Pre-Training with Indirect Data Poisoning

The pre-training of large language models LLMs relies on massive text datasets sourced from diverse and difficult-to-curate origins. Although membership inference attacks and hidden canaries have been explored to trace data usage, such methods rely on memorization of training data, which LM...

ExtendAttack: Attacking Servers of LRMs via Extending Reasoning

Large Reasoning Models LRMs have demonstrated promising performance in complex tasks. However, the resource-consuming reasoning processes may be exploited by attackers to maliciously occupy the resources of the servers, leading to a crash, like the DDoS attack in cyber. To this end, we propose a...

Doppelgänger Method: Breaking Role Consistency in LLM Agent via Prompt-based Transferable Adversarial Attack

Since the advent of large language models, prompt engineering now enables the rapid, low-effort creation of diverse autonomous agents that are already in widespread use. Yet this convenience raises urgent concerns about the safety, robustness, and behavioral consistency of the underlying prompts,...

FORTRESS: Frontier Risk Evaluation for National Security and Public Safety

The rapid advancement of large language models LLMs introduces dual-use capabilities that could both threaten and bolster national security and public safety NSPS. Models implement safeguards to protect against potential misuse relevant to NSPS and allow for benign users to receive helpful...

Weakest Link in the Chain: Security Vulnerabilities in Advanced Reasoning Models

The introduction of advanced reasoning capabilities have improved the problem-solving performance of large language models, particularly on math and coding benchmarks. However, it remains unclear whether these reasoning models are more or less vulnerable to adversarial prompt attacks than their...

CVE-2025-5255 TCC Bypass via Dylib Injection in Phoenix Code

The Phoenix Code's configuration on macOS, specifically the presence of entitlements: "com.apple.security.cs.allow-dyld-environment-variables" and "com.apple.security.cs.disable-library-validation" allows for Dynamic Library Dylib injection. A local attacker with unprivileged access can use...

IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems

The rapid advancement of Large Language Models LLMs has led to the emergence of Multi-Agent Systems MAS to perform complex tasks through collaboration. However, the intricate nature of MAS, including their architecture and agent interactions, raises significant concerns regarding intellectual...

Astra Linux – Vulnerability in Chromium

Inappropriate implementation in Permission Prompts in Google Chrome prior to 134.0.6998.35 allowed an attacker who convinced a user to install a malicious extension to perform UI spoofing via a crafted Chrome Extension. Chromium security severity: Low...

TencentOS Server 4: thunderbird (TSSA-2024:0668)

The version of Tencent Linux installed on the remote TencentOS Server 4 host is prior to tested version. It is, therefore, affected by multiple vulnerabilities as referenced in the TSSA-2024:0668 advisory. Package updates are available for TencentOS Server 4 that fix the following vulnerabilities...

SOSBENCH: Benchmarking Safety Alignment on Scientific Knowledge

Large language models LLMs exhibit advancing capabilities in complex tasks, such as reasoning and graduate-level question answering, yet their resilience against misuse, particularly involving scientifically sophisticated risks, remains underexplored. Existing safety benchmarks typically focus...

Chain-Of-Code Collapse: Reasoning Failures in LLMs Via Adversarial Prompting in Code Generation

Large Language Models LLMs have achieved remarkable success in tasks requiring complex reasoning, such as code generation, mathematical problem solving, and algorithmic synthesis -- especially when aided by reasoning tokens and Chain-of-Thought prompting. Yet, a core question remains: do these...

GenBreak: Red Teaming Text-To-Image Generators Using Large Language Models

Text-to-image T2I models such as Stable Diffusion have advanced rapidly and are now widely used in content creation. However, these models can be misused to generate harmful content, including nudity or violence, posing significant safety risks. While most platforms employ content moderation...

LLMs Caught in the Crossfire: Malware Requests and Jailbreak Challenges

The widespread adoption of Large Language Models LLMs has heightened concerns about their security, particularly their vulnerability to jailbreak attacks that leverage crafted prompts to generate malicious outputs. While prior research has been conducted on general security capabilities of LLMs,...

SUSE-SU-2025:20394-1 Security update for less

This update for less fixes the following issues: - Updated to version 668 Fixed crash when using --header on command line Fixed possible crash when scrolling left/right or toggling -S Fixed bug when using stop in a lesskey file Fixed bug when using --shift or --match-shift on command line with a...

CVE-2025-5018

The Hive Support plugin for WordPress is vulnerable to unauthorized access and modification of data due to a missing capability check on the hsupdateaichatsettings and hivelitesupportgetallbinbox functions in all versions up to, and including, 1.2.5. This makes it possible for authenticated...

Privacy and Security Threat for OpenAI GPTs

Large language models LLMs demonstrate powerful information handling capabilities and are widely integrated into chatbot applications. OpenAI provides a platform for developers to construct custom GPTs, extending ChatGPT's functions and integrating external services. Since its release in November...

ReGA: Representation-Guided Abstraction for Model-Based Safeguarding of LLMs

Large Language Models LLMs have achieved significant success in various tasks, yet concerns about their safety and security have emerged. In particular, they pose risks in generating harmful content and vulnerability to jailbreaking attacks. To analyze and monitor machine learning models,...

When GPT Spills the Tea: Comprehensive Assessment of Knowledge File Leakage in GPTs

Knowledge files have been widely used in large language model LLM agents, such as GPTs, to improve response quality. However, concerns about the potential leakage of knowledge files have grown significantly. Existing studies demonstrate that adversarial prompts can induce GPTs to leak knowledge...

Fooling the Watchers: Breaking AIGC Detectors Via Semantic Prompt Attacks

The rise of text-to-image T2I models has enabled the synthesis of photorealistic human portraits, raising serious concerns about identity misuse and the robustness of AIGC detectors. In this work, we propose an automated adversarial prompt generation framework that leverages a grammar tree...

Privacy-Preserving Prompt Personalization in Federated Learning for Multimodal Large Language Models

Prompt learning is a crucial technique for adapting pre-trained multimodal language models MLLMs to user tasks. Federated prompt personalization FPP is further developed to address data heterogeneity and local overfitting, however, it exposes personalized prompts - valuable intellectual assets - ...