Enhancing Targeted Adversarial Attacks on Large Vision-Language Models through Intermediate Projector Guidance

Targeted adversarial attacks are essential for proactively identifying security flaws in Vision-Language Models before real-world deployment. However, current methods perturb images to maximize global similarity with the target text or reference image at the encoder level, collapsing rich visual...

Adversarial Attacks on VQA-NLE: Exposing and Alleviating Inconsistencies in Visual Question Answering Explanations

Natural language explanations in visual question answering VQA-NLE aim to make black-box models more transparent by elucidating their decision-making processes. However, we find that existing VQA-NLE systems can produce inconsistent explanations and reach conclusions without genuinely understandi...

MM-AttacKG: a Multimodal Approach to Attack Graph Construction with Large Language Models

Cyber Threat Intelligence CTI parsing aims to extract key threat information from massive data, transform it into actionable intelligence, enhance threat detection and defense efficiency, including attack graph construction, intelligence fusion and indicator extraction. Among these research topic...

CVE-2019-17191

The Signal Private Messenger application before 4.47.7 for Android allows a caller to force a call to be answered, without callee user interaction, via a connect message. The existence of the call is noticeable to the callee; however, the audio channel may be open before the callee can block...

Apple iOS和Apple iPadOS 安全漏洞

Apple iOS and Apple iPadOS are products of Apple Inc. Apple iOS is an operating system developed for mobile devices, and Apple iPadOS is an operating system for iPad tablets. A security vulnerability exists in Apple iOS and Apple iPadOS that stems from an insufficient restriction of options on...

Building Trustworthy Multimodal AI: a Review of Fairness, Transparency, and Ethics in Vision-Language Tasks

Objective: This review explores the trustworthiness of multimodal artificial intelligence AI systems, specifically focusing on vision-language tasks. It addresses critical challenges related to fairness, transparency, and ethical implications in these systems, providing a comparative analysis of...

Supercharging Your AI Applications with Spring AI Advisors

In the rapidly evolving world of artificial intelligence, developers are constantly seeking ways to enhance their AI applications. Spring AI, a Java framework for building AI-powered applications, has introduced a powerful feature: the Spring AI Advisors. The advisors can supercharge your AI...

Weak Password Vulnerability in Hybrid Alarm Receiver at Honeywell (China) Co.

Ltd. is part of Honeywell International, a diversified, high-tech advanced manufacturing company with a turnover of $38 billion. A weak password vulnerability exists in Honeywell China Limited's Hybrid Answering Machine, which can be exploited by attackers to obtain sensitive information...

CVE-2020-16169

Authentication Bypass Using an Alternate Path or Channel in temi Robox OS prior to120, temi Android app up to 1.3.7931 allows remote attackers to gain elevated privileges on the temi and have it automatically answer the attacker's calls, granting audio, video, and motor control via unspecified...

DEBIAN-CVE-2019-16375

An issue was discovered in Open Ticket Request System OTRS 7.0.x through 7.0.11, and Community Edition 5.0.x through 5.0.37 and 6.0.x through 6.0.22. An attacker who is logged in as an agent or customer user with appropriate permissions can create a carefully crafted string containing malicious...

CVE-2019-17191

The CVE-2019-17191 entry concerns the Signal Private Messenger app for Android prior to version 4.47.7. The underling issue is a flow that lets a caller force a call to be answered via a connect message without recipient interaction. The callee is aware a call exists, but the audio channel may be...

Unauthorised Answering Attempts

Moodle is vulnerable to unauthorized answering attempts. The vulnerability is because it does not restrict the maximum number of attempts for the completed and graded lesson activity, allowing the students to repeat their attempts to answer the questions...

Tipask question Answering System 1 2 injection package-vulnerability warning-the black bar safety net

Participate in a period of public test of the time, manufacturers use this system, download down looked, found injected into a large heap. Since the program after the end of the sql statement in many places without using the single quotes the parameter is enclosed, so the user input of the check...

oask question Answering System v2. 0 0day-vulnerability warning-the black bar safety net

Post a oask question and answer system 0day latest version http://127.0.0.1/admin/index.asp The background configuration file, in the Administrator's mailbox of written word a"%%eval request"a"'...