Spa-VLM: Stealthy Poisoning Attacks on RAG-Based VLM

With the rapid development of the Vision-Language Model VLM, significant progress has been made in Visual Question Answering VQA tasks. However, existing VLM often generate inaccurate answers due to a lack of up-to-date knowledge. To address this issue, recent research has introduced...

A Survey on the Safety and Security Threats of Computer-Using Agents: JARVIS or Ultron?

Recently, AI-driven interactions with computing devices have advanced from basic prototype tools to sophisticated, LLM-based systems that emulate human-like operations in graphical user interfaces. We are now witnessing the emergence of \emphComputer-Using Agents CUAs, capable of autonomously...

Benchmarking Poisoning Attacks against Retrieval-Augmented Generation

Retrieval-Augmented Generation RAG has proven effective in mitigating hallucinations in large language models by incorporating external knowledge during inference. However, this integration introduces new security vulnerabilities, particularly to poisoning attacks. Although prior work has explore...

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...

Beyond Text: Unveiling Privacy Vulnerabilities in Multi-Modal Retrieval-Augmented Generation

Multimodal Retrieval-Augmented Generation MRAG systems enhance LMMs by integrating external multimodal databases, but introduce unexplored privacy vulnerabilities. While text-based RAG privacy risks have been studied, multimodal data presents unique challenges. We provide the first systematic...

CVE-2025-4897

A vulnerability was found in Tenda A15 15.13.07.09/15.13.07.13. It has been classified as critical. This affects an unknown part of the file /goform/multimodalAdd of the component HTTP POST Request Handler. The manipulation leads to buffer overflow. It is possible to initiate the attack remotely...

Tenda A15 安全漏洞

The Tenda A15 is a dual-band gigabit wireless router. A buffer overflow vulnerability exists in the Tenda A15. The vulnerability stems from improper handling of the /goform/multimodalAdd file by the HTTP POST request processor component and can be exploited by an attacker to remotely execute...

AC-LoRA: (Almost) Training-Free Access Control-Aware Multi-Modal LLMs

Corporate LLMs are gaining traction for efficient knowledge dissemination and management within organizations. However, as current LLMs are vulnerable to leaking sensitive information, it has proven difficult to apply them in settings where strict access control is necessary. To this end, we desi...

WhatsAI: Transforming Meta Ray-Bans into an Extensible Generative AI Platform for Accessibility

Multi-modal generative AI models integrated into wearable devices have shown significant promise in enhancing the accessibility of visual information for blind or visually impaired BVI individuals, as evidenced by the rapid uptake of Meta Ray-Bans among BVI users. However, the proprietary nature ...

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...

CVE-2025-46560 vLLM phi4mm: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.8.0 and prior to 0.8.5 are affected by a critical performance vulnerability in the input preprocessing logic of the multimodal tokenizer. The code dynamically replaces placeholder tokens...

CVE-2025-46560 vLLM phi4mm: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.8.0 and prior to 0.8.5 are affected by a critical performance vulnerability in the input preprocessing logic of the multimodal tokenizer. The code dynamically replaces placeholder tokens...

CVE-2025-46560: Inefficient Regular Expression Complexity

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.8.0 and prior to 0.8.5 are affected by a critical performance vulnerability in the input preprocessing logic of the multimodal tokenizer. The code dynamically replaces placeholder tokens...

GHSA-VC6M-HM49-G9QG vLLM: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

Summary A critical performance vulnerability has been identified in the input preprocessing logic of the multimodal tokenizer. The code dynamically replaces placeholder tokens e.g., , with repeated tokens based on precomputed lengths. Due to ​​inefficient list concatenation operations​​, the...

AGATE: Stealthy Black-Box Watermarking for Multimodal Model Copyright Protection

Recent advancement in large-scale Artificial Intelligence AI models offering multimodal services have become foundational in AI systems, making them prime targets for model theft. Existing methods select Out-of-Distribution OoD data as backdoor watermarks and retrain the original model for...

T2VShield: Model-Agnostic Jailbreak Defense for Text-To-Video Models

The rapid development of generative artificial intelligence has made text to video models essential for building future multimodal world simulators. However, these models remain vulnerable to jailbreak attacks, where specially crafted prompts bypass safety mechanisms and lead to the generation of...

Semantic-Aware Contrastive Fine-Tuning: Boosting Multimodal Malware Classification with Discriminative Embeddings

The rapid evolution of malware variants requires robust classification methods to enhance cybersecurity. While Large Language Models LLMs offer potential for generating malware descriptions to aid family classification, their utility is limited by semantic embedding overlaps and misalignment with...

IoT-AMLHP: Aligned Multimodal Learning of Header-Payload Representations for Resource-Efficient Malicious IoT Traffic Classification

Traffic classification is crucial for securing Internet of Things IoT networks. Deep learning-based methods can autonomously extract latent patterns from massive network traffic, demonstrating significant potential for IoT traffic classification tasks. However, the limited computational and spati...

On the Feasibility of Using MultiModal LLMs to Execute AR Social Engineering Attacks

Augmented Reality AR and Multimodal Large Language Models LLMs are rapidly evolving, providing unprecedented capabilities for human-computer interaction. However, their integration introduces a new attack surface for social engineering. In this paper, we systematically investigate the feasibility...

The vulnerability of the check_access() function in the system for launching and managing large language multimodal systems (LoLLMS) allows a perpetrator to gain access to read, modify, or delete data, or to cause service failures.

The vulnerability of the checkaccess function in the system for launching and managing large language multimodal systems LoLLMS is related to deficiencies in the authentication process. Exploiting this vulnerability could allow an attacker to gain read, modify, or delete access to data, or to cau...