Low Rank Comes with Low Security: Gradient Assembly Poisoning Attacks against Distributed LoRA-Based LLM Systems

Low-Rank Adaptation LoRA has become a popular solution for fine-tuning large language models LLMs in federated settings, dramatically reducing update costs by introducing trainable low-rank matrices. However, when integrated with frameworks like FedIT, LoRA introduces a critical vulnerability:...

Design and Detection of Covert Man-In-The-Middle Cyberattacks on Water Treatment Plants

Cyberattacks targeting critical infrastructures, such as water treatment facilities, represent significant threats to public health, safety, and the environment. This paper introduces a systematic approach for modeling and assessing covert man-in-the-middle MitM attacks that leverage system...

HAMLOCK: HArdware-Model LOgically Combined AttacK

The growing use of third-party hardware accelerators e.g., FPGAs, ASICs for deep neural networks DNNs introduces new security vulnerabilities. Conventional model-level backdoor attacks, which only poison a model's weights to misclassify inputs with a specific trigger, are often detectable because...

SilentStriker: toward Stealthy Bit-Flip Attacks on Large Language Models

The rapid adoption of large language models LLMs in critical domains has spurred extensive research into their security issues. While input manipulation attacks e.g., prompt injection have been well studied, Bit-Flip Attacks BFAs -- which exploit hardware vulnerabilities to corrupt model paramete...

When Forgetting Triggers Backdoors: a Clean Unlearning Attack

Machine unlearning has emerged as a key component in ensuring Right to be Forgotten, enabling the removal of specific data points from trained models. However, even when the unlearning is performed without poisoning the forget-set clean unlearning, it can be exploited for stealthy attacks that...

Keeping an Eye on LLM Unlearning: the Hidden Risk and Remedy

Although Large Language Models LLMs have demonstrated impressive capabilities across a wide range of tasks, growing concerns have emerged over the misuse of sensitive, copyrighted, or harmful data during training. To address these concerns, unlearning techniques have been developed to remove the...

Chinese Hackers Breach Asian Telecom, Remain Undetected for Over 4 Years

A major telecommunications company located in Asia was allegedly breached by Chinese state-sponsored hackers who spent over four years inside its systems, according to a new report from incident response firm Sygnia. The cybersecurity company is tracking the activity under the name Weaver Ant ,...

PROPagate Code Injection Seen in the Wild

Last year, researchers wrote about a new Windows code injection technique called PROPagate. Last week, it was first seen in malware: This technique abuses the SetWindowsSubclass function -- a process used to install or update subclass windows running on the system -- and can be used to modify the...

Cross-Site Scripting (CSS) in Hosting Controller All Version and hot fix it hehe ;)

Cross-Site Scripting CSS in Hosting Controller All Version and hot fix it hehe ; [email protected] Ashiyane Security Team SegmentationFault Group iam sun-os hehe :P Greetz to : S4P0 - failed - Xsupr3mo - Status-x - Stealh Greetz to : berhooz - nima - ehsan - Mehrtash and...