Efficient Decoding Methods for Language Models on Encrypted Data

Large language models LLMs power modern AI applications, but processing sensitive data on untrusted servers raises privacy concerns. Homomorphic encryption HE enables computation on encrypted data for secure inference. However, neural text generation requires decoding methods like argmax and...

GHSA-6VM5-6JV9-RJPJ MONAI: Unsafe torch usage may lead to arbitrary code execution

Summary In modeldict = torch.loadfullpath, maplocation=torch.devicedevice, weightsonly=True in monai/bundle/scripts.py , weightsonly=True is loaded securely. However, insecure loading methods still exist elsewhere in the project, such as when loading checkpoints. This is a common practice when...

CVE-2025-54084

OS Command 'OS Command Injection' vulnerability in Calix GigaCenter ONT Quantenna SoC modules allows authenticated attackers with 'super' user credentials to execute arbitrary OS commands through improper input validation, potentially leading to full system compromise.This issue affects GigaCente...

CVE-2025-54084 Calix Gigacenter ONT - Command Injection

OS Command 'OS Command Injection' vulnerability in Calix GigaCenter ONT Quantenna SoC modules allows authenticated attackers with 'super' user credentials to execute arbitrary OS commands through improper input validation, potentially leading to full system compromise.This issue affects GigaCente...

CVE-2025-53913

CVE-2025-53913 affects Calix GigaCenter ONT (Quantenna SoC modules). Affected models: 844E, 844G, 844GE, 854GE, 812G, 813G, 818G. Issue: Excessive privileges enabling privilege abuse in base system. Documented impact includes high confidentiality, integrity, and availability implications, with ph...

CVE-2025-53913 Calix GigaCenter ONT (Quantenna SoC) - Excessive Privileges

Excessive Privileges vulnerability in Calix GigaCenter ONT Quantenna SoC modules allows Privilege Abuse.This issue affects GigaCenter ONT: 844E, 844G, 844GE, 854GE, 812G, 813G, 818G...

PYSEC-2025-141

MONAI Medical Open Network for AI is an AI toolkit for health care imaging. In versions up to and including 1.5.0, in modeldict = torch.loadfullpath, maplocation=torch.devicedevice, weightsonly=True in monai/bundle/scripts.py , weightsonly=True is loaded securely. However, insecure loading method...

Breaking Android with AI: a Deep Dive into LLM-Powered Exploitation

The rapid evolution of Artificial Intelligence AI and Large Language Models LLMs has opened up new opportunities in the area of cybersecurity, especially in the exploitation automation landscape and penetration testing. This study explores Android penetration testing automation using LLM-based...

PT-2025-36947

Name of the Vulnerable Software and Affected Versions: Calix GigaCenter ONT versions 844E Calix GigaCenter ONT versions 844G Calix GigaCenter ONT versions 844GE Calix GigaCenter ONT versions 854GE Calix GigaCenter ONT versions 812G Calix GigaCenter ONT versions 813G Calix GigaCenter ONT versions...

PatchSeeker: Mapping NVD Records to Their Vulnerability-Fixing Commits with LLM Generated Commits and Embeddings

Software vulnerabilities pose serious risks to modern software ecosystems. While the National Vulnerability Database NVD is the authoritative source for cataloging these vulnerabilities, it often lacks explicit links to the corresponding Vulnerability-Fixing Commits VFCs. VFCs encode precise code...

Guided Reasoning in LLM-Driven Penetration Testing Using Structured Attack Trees

Recent advances in Large Language Models LLMs have driven interest in automating cybersecurity penetration testing workflows, offering the promise of faster and more consistent vulnerability assessment for enterprise systems. Existing LLM agents for penetration testing primarily rely on self-guid...

Baicells多款产品 信息泄露漏洞

Baicells Nova 436Q and others are products of Baicells, Inc.Baicells Nova 436Q is an advanced dual-carrier outdoor eNodeB eNB. Baicells Nova 243 and others are products of the company.Baicells Nova 243 is a miniature base station.Baicells NEUTRINO430 is an LTE base station. An information...

Calix GigaCenter ONT Series 安全漏洞

Calix GigaCenter ONT Series is a series of optical network terminals from Calix USA. A security vulnerability exists in the Calix GigaCenter ONT Series that originates from a common user being able to abuse elevated privileges. The following products and versions are affected: GigaCenter ONT 844E...

LLMs in Cybersecurity: Friend or Foe in the Human Decision Loop?

Large Language Models LLMs are transforming human decision-making by acting as cognitive collaborators. Yet, this promise comes with a paradox: while LLMs can improve accuracy, they may also erode independent reasoning, promote over-reliance and homogenize decisions. In this paper, we investigate...

PT-2025-36532

Name of the Vulnerable Software and Affected Versions: MONAI versions up to and including 1.5.0 Description: MONAI is an AI toolkit for health care imaging. The software contains insecure model loading methods that can trigger a deserialization vulnerability, potentially leading to code execution...

Mask-GCG: Are All Tokens in Adversarial Suffixes Necessary for Jailbreak Attacks?

Jailbreak attacks on Large Language Models LLMs have demonstrated various successful methods whereby attackers manipulate models into generating harmful responses that they are designed to avoid. Among these, Greedy Coordinate Gradient GCG has emerged as a general and effective approach that...

Multimodal Prompt Injection Attacks: Risks and Defenses for Modern LLMs

Large Language Models LLMs have seen rapid adoption in recent years, with industries increasingly relying on them to maintain a competitive advantage. These models excel at interpreting user instructions and generating human-like responses, leading to their integration across diverse domains,...

CVE-2025-35452

PTZOptics and ValueHD-based PTZ cameras are affected by an insufficient authentication issue affecting the administrative web interface. The vulnerability stems from default, shared credentials and weak access controls for the /cgi-bin/param.cgi path, enabling remote, unauthenticated access to se...

CVE-2025-35452 Pan-Tilt-Zoom cameras default administrative credentials for web interface

PTZOptics and possibly other ValueHD-based pan-tilt-zoom cameras use default, shared credentials for the administrative web interface...

Behind the Mask: Benchmarking Camouflaged Jailbreaks in Large Language Models

Large Language Models LLMs are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety mechanisms. Unlike overt attacks, these subtle prompts exploit...