vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions

Summary Users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimension mismatch that results in an unhandled runtime error, leading to complete server termination. Details T...

EUVD-2026-1865

vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions...

CVE-2026-21219

Use after free in Inbox COM Objects allows an unauthorized attacker to execute code locally...

CVE-2026-21219 Inbox COM Objects (Global Memory) Remote Code Execution Vulnerability

...

CVE-2026-21219 Inbox COM Objects (Global Memory) Remote Code Execution Vulnerability

...

MINI-43RH-VMRH-47M9

Bulletin has no description...

CVE-2026-22755

Improper Neutralization of Special Elements used in a Command 'Command Injection' vulnerability in Vivotek Affected device model numbers are FD8365, FD8365v2, FD9165, FD9171, FD9187, FD9189, FD9365, FD9371, FD9381, FD9387, FD9389, FD9391,FE9180,FE9181, FE9191, FE9381, FE9382, FE9391, FE9582,...

CVE-2026-0889

Denial-of-service in the DOM: Service Workers component. This vulnerability was fixed in Firefox 147 and Thunderbird 147...

MINI-47WF-2C4P-2M3H

Bulletin has no description...

CVE-2026-0890 Spoofing issue in the DOM: Copy & Paste and Drag & Drop component

Spoofing issue in the DOM: Copy & Paste and Drag & Drop component. This vulnerability was fixed in Firefox 147, Firefox ESR 140.7, Thunderbird 147, and Thunderbird 140.7...

CVE-2026-0889

Denial-of-service in the DOM: Service Workers component. This vulnerability affects Firefox 147 and Thunderbird 147...

CVE-2026-0877

Mitigation bypass in the DOM: Security component. This vulnerability was fixed in Firefox 147, Firefox ESR 115.32, Firefox ESR 140.7, Thunderbird 147, and Thunderbird 140.7...

BIT-GITLAB-2025-13772 Missing Authorization in GitLab

GitLab has remediated an issue in GitLab EE affecting all versions from 18.4 before 18.5.5, 18.6 before 18.6.3, and 18.7 before 18.7.1 that could have allowed an authenticated user to access and utilize AI model settings from unauthorized namespaces by manipulating namespace identifiers in API...

Authentication Bypass

Ollama is vulnerable to an Authentication Bypass. The vulnerability is due to where critical model management APIs are exposed without access controls, allowing remote attackers to perform unauthorized operations without authentication...

Integer Overflow lead to DOS in handling Accept-Encoding header in API /v2/models/<model-name>/generate

This report is not public...

Proactively Detecting Threats: A Novel Approach Using LLMs

Enterprise security faces escalating threats from sophisticated malware, compounded by expanding digital operations. This paper presents the first systematic evaluation of large language models LLMs to proactively identify indicators of compromise IOCs from unstructured web-based threat...

PT-2026-3193

Name of the Vulnerable Software and Affected Versions AVEVA Process Optimization affected versions not specified Description A flaw exists that could allow an attacker to execute code remotely on the system with operating system level privileges through the taoimr service. Successful exploitation...

PT-2026-3195

Name of the Vulnerable Software and Affected Versions versions prior to 2025 affected versions not specified Description An authenticated user with standard operating system privileges could modify TCL Macro scripts. Successful exploitation may lead to privilege escalation to the operating system...

Integrating APK Image and Text Data for Enhanced Threat Detection: A Multimodal Deep Learning Approach to Android Malware

As zero-day Android malware attacks grow more sophisticated, recent research highlights the effectiveness of using image-based representations of malware bytecode to detect previously unseen threats. However, existing studies often overlook how image type and resolution affect detection and ignor...

A Decompilation-Driven Framework for Malware Detection with Large Language Models

The parallel evolution of Large Language Models LLMs with advanced code-understanding capabilities and the increasing sophistication of malware presents a new frontier for cybersecurity research. This paper evaluates the efficacy of state-of-the-art LLMs in classifying executable code as either...