CVE-2012-3290

Multiple unspecified vulnerabilities in Google Chrome before 20.0.1132.22 on the Acer AC700; Samsung Series 5, 5 550, and Chromebox 3; and Cr-48 Chromebook platforms have unknown impact and attack vectors...

CVE-2018-21229

Certain NETGEAR devices are affected by incorrect configuration of security settings. This affects R7500v2 before 1.0.3.20, R7800 before 1.0.2.38, WN3000RPv3 before 1.0.2.50, WNDR4300v2 before 1.0.0.50, and WNDR4500v3 before 1.0.0.50...

CVE-2018-21204

Certain NETGEAR devices are affected by a stack-based buffer overflow by an unauthenticated attacker. This affects D7800 before 1.0.1.30, R6100 before 1.0.1.20, R7500 before 1.0.0.118, R7500v2 before 1.0.3.24, R7800 before 1.0.2.40, R9000 before 1.0.2.52, WNDR3700v4 before 1.0.2.96, WNDR4300 befo...

CVE-2018-21219

Certain NETGEAR devices are affected by a buffer overflow by an unauthenticated attacker. This affects D3600 before 1.0.0.67, D6000 before 1.0.0.67, D6100 before 1.0.0.56, D7800 before 1.0.1.30, R6100 before 1.0.1.20, R7500 before 1.0.0.118, R7500v2 before 1.0.3.24, R9000 before 1.0.2.52,...

CVE-2018-11967

Signature verification of the skel library could potentially be disabled as the memory region on the remote subsystem in which the library is loaded is allocated from userspace currently in Snapdragon Auto, Snapdragon Compute, Snapdragon Connectivity, Snapdragon Consumer IOT, Snapdragon Industria...

CVE-2013-4829

HP LaserJet M4555, M525, and M725; LaserJet flow MFP M525c; LaserJet Enterprise color flow MFP M575c; Color LaserJet CM4540, M575, and M775; and ScanJet Enterprise 8500fn1 FutureSmart devices allow local users to read images of arbitrary scanned documents via unspecified vectors...

CVE-2017-18827

Certain NETGEAR devices are affected by stored XSS. This affects M4300-28G before 12.0.2.15, M4300-52G before 12.0.2.15, M4300-28G-POE+ before 12.0.2.15, M4300-52G-POE+ before 12.0.2.15, M4300-8X8F before 12.0.2.15, M4300-12X12F before 12.0.2.15, M4300-24X24F before 12.0.2.15, M4300-24X before...

Mitigating Fine-Tuning Risks in LLMs Via Safety-Aware Probing Optimization

The significant progress of large language models LLMs has led to remarkable achievements across numerous applications. However, their ability to generate harmful content has sparked substantial safety concerns. Despite the implementation of safety alignment techniques during the pre-training...

When Safety Detectors Aren'T Enough: a Stealthy and Effective Jailbreak Attack on LLMs Via Steganographic Techniques

Jailbreak attacks pose a serious threat to large language models LLMs by bypassing built-in safety mechanisms and leading to harmful outputs. Studying these attacks is crucial for identifying vulnerabilities and improving model security. This paper presents a systematic survey of jailbreak method...

Backdoor Cleaning without External Guidance in MLLM Fine-Tuning

Multimodal Large Language Models MLLMs are increasingly deployed in fine-tuning-as-a-service FTaaS settings, where user-submitted datasets adapt general-purpose models to downstream tasks. This flexibility, however, introduces serious security risks, as malicious fine-tuning can implant backdoors...

DuFFin: a Dual-Level Fingerprinting Framework for LLMs IP Protection

Whitepaper called DuFFin: A Dual-Level Fingerprinting Framework For LLMs IP Protection...

Password Strength Detection Via Machine Learning: Analysis, Modeling, and Evaluation

As network security issues continue gaining prominence, password security has become crucial in safeguarding personal information and network systems. This study first introduces various methods for system password cracking, outlines password defense strategies, and discusses the application of...

Harry Potter Is Still Here! Probing Knowledge Leakage in Targeted Unlearned Large Language Models Via Automated Adversarial Prompting

This work presents LURK Latent UnleaRned Knowledge, a novel framework that probes for hidden retained knowledge in unlearned LLMs through adversarial suffix prompting. LURK automatically generates adversarial prompt suffixes designed to elicit residual knowledge about the Harry Potter domain, a...

Advancing Security with Digital Twins: a Comprehensive Survey

The proliferation of electronic devices has greatly transformed every aspect of human life, such as communication, healthcare, transportation, and energy. Unfortunately, the global electronics supply chain is vulnerable to various attacks, including piracy of intellectual properties, tampering,...

CAIN: Hijacking LLM-Humans Conversations Via a Two-Stage Malicious System Prompt Generation and Refining Framework

Large language models LLMs have advanced many applications, but are also known to be vulnerable to adversarial attacks. In this work, we introduce a novel security threat: hijacking AI-human conversations by manipulating LLMs' system prompts to produce malicious answers only to specific targeted...

CoTSRF: Utilize Chain of Thought As Stealthy and Robust Fingerprint of Large Language Models

Despite providing superior performance, open-source large language models LLMs are vulnerable to abusive usage. To address this issue, recent works propose LLM fingerprinting methods to identify the specific source LLMs behind suspect applications. However, these methods fail to provide stealthy...

MTSA: Multi-Turn Safety Alignment for LLMs through Multi-Round Red-Teaming

Whitepaper called MTSA: Multi-Turn Safety Alignment For LLMs Through Multi-Round Red-Teaming...

Unlearning Isn'T Deletion: Investigating Reversibility of Machine Unlearning in LLMs

Unlearning in large language models LLMs is intended to remove the influence of specific data, yet current evaluations rely heavily on token-level metrics such as accuracy and perplexity. We show that these metrics can be misleading: models often appear to forget, but their original behavior can ...

Blind Spot Navigation: Evolutionary Discovery of Sensitive Semantic Concepts for LVLMs

Whitepaper called Blind Spot Navigation: Evolutionary Discovery Of Sensitive Semantic Concepts For LVLMs...

Scalable Defense against In-The-Wild Jailbreaking Attacks with Safety Context Retrieval

Large Language Models LLMs are known to be vulnerable to jailbreaking attacks, wherein adversaries exploit carefully engineered prompts to induce harmful or unethical responses. Such threats have raised critical concerns about the safety and reliability of LLMs in real-world deployment. While...