llama.cpp 安全漏洞

llama.cpp is a multimodal model by the individual developer Georgi Gerganov. A security vulnerability exists in versions of llama.cpp prior to b5721, which stems from the presence of signed and unsigned integer overflows in the tokenizer implementation, which could lead to a heap overflow...

Pushing the Limits of Safety: a Technical Report on the ATLAS Challenge 2025

Multimodal Large Language Models MLLMs have enabled transformative advancements across diverse applications but remain susceptible to safety threats, especially jailbreak attacks that induce harmful outputs. To systematically evaluate and improve their safety, we organized the Adversarial Testing...

SmartHome-Bench: a Comprehensive Benchmark for Video Anomaly Detection in Smart Homes Using Multi-Modal Large Language Models

Video anomaly detection VAD is essential for enhancing safety and security by identifying unusual events across different environments. Existing VAD benchmarks, however, are primarily designed for general-purpose scenarios, neglecting the specific characteristics of smart home applications. To...

Investigating Vulnerabilities and Defenses against Audio-Visual Attacks: a Comprehensive Survey Emphasizing Multimodal Models

Multimodal large language models MLLMs, which bridge the gap between audio-visual and natural language processing, achieve state-of-the-art performance on several audio-visual tasks. Despite the superior performance of MLLMs, the scarcity of high-quality audio-visual training data and computation...

DinoCompanion: an Attachment-Theory Informed Multimodal Robot for Emotionally Responsive Child-AI Interaction

Children's emotional development fundamentally relies on secure attachment relationships, yet current AI companions lack the theoretical foundation to provide developmentally appropriate emotional support. We introduce DinoCompanion, the first attachment-theory-grounded multimodal robot for...

InverTune: Removing Backdoors from Multimodal Contrastive Learning Models Via Trigger Inversion and Activation Tuning

Multimodal contrastive learning models like CLIP have demonstrated remarkable vision-language alignment capabilities, yet their vulnerability to backdoor attacks poses critical security risks. Attackers can implant latent triggers that persist through downstream tasks, enabling malicious control ...

LingoLoop Attack: Trapping MLLMs via Linguistic Context and State Entrapment into Endless Loops

Multimodal Large Language Models MLLMs have shown great promise but require substantial computational resources during inference. Attackers can exploit this by inducing excessive output, leading to resource exhaustion and service degradation. Prior energy-latency attacks aim to increase generatio...

From LLMs to MLLMs to Agents: a Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystem

Large language models LLMs are rapidly evolving from single-modal systems to multimodal LLMs and intelligent agents, significantly expanding their capabilities while introducing increasingly severe security risks. This paper presents a systematic survey of the growing complexity of jailbreak...

Foundation of Affective Computing and Interaction

This book provides a comprehensive exploration of affective computing and human-computer interaction technologies. It begins with the historical development and basic concepts of human-computer interaction, delving into the technical frameworks and practical applications of emotional computing,...

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...

Safeguarding Multimodal Knowledge Copyright in the RAG-As-A-Service Environment

As Retrieval-Augmented Generation RAG evolves into service-oriented platforms Rag-as-a-Service with shared knowledge bases, protecting the copyright of contributed data becomes essential. Existing watermarking methods in RAG focus solely on textual knowledge, leaving image knowledge unprotected. ...

Doxing Via the Lens: Revealing Location-Related Privacy Leakage on Multi-Modal Large Reasoning Models

Recent advances in multi-modal large reasoning models MLRMs have shown significant ability to interpret complex visual content. While these models enable impressive reasoning capabilities, they also introduce novel and underexplored privacy risks. In this paper, we identify a novel category of...

Dual-Priv Pruning : Efficient Differential Private Fine-Tuning in Multimodal Large Language Models

Differential Privacy DP is a widely adopted technique, valued for its effectiveness in protecting the privacy of task-specific datasets, making it a critical tool for large language models. However, its effectiveness in Multimodal Large Language Models MLLMs remains uncertain. Applying Differenti...

Fuse and Federate: Enhancing EV Charging Station Security with Multimodal Fusion and Federated Learning

The rapid global adoption of electric vehicles EVs has established electric vehicle supply equipment EVSE as a critical component of smart grid infrastructure. While essential for ensuring reliable energy delivery and accessibility, EVSE systems face significant cybersecurity challenges, includin...

VPI-Bench: Visual Prompt Injection Attacks for Computer-Use Agents

Computer-Use Agents CUAs with full system access enable powerful task automation but pose significant security and privacy risks due to their ability to manipulate files, access user data, and execute arbitrary commands. While prior work has focused on browser-based agents and HTML-level attacks,...

Align Is Not Enough: Multimodal Universal Jailbreak Attack against Multimodal Large Language Models

Large Language Models LLMs have evolved into Multimodal Large Language Models MLLMs, significantly enhancing their capabilities by integrating visual information and other types, thus aligning more closely with the nature of human intelligence, which processes a variety of data forms beyond just...

Con Instruction: Universal Jailbreaking of Multimodal Large Language Models Via Non-Textual Modalities

Existing attacks against multimodal language models MLLMs primarily communicate instructions through text accompanied by adversarial images. In contrast, we exploit the capabilities of MLLMs to interpret non-textual instructions, specifically, adversarial images or audio generated by our novel...

CVE-2025-46722: Improper Validation of Consistency within Input

vLLM is an inference and serving engine for large language models LLMs. In versions starting from 0.7.0 to before 0.9.0, in the file vllm/multimodal/hasher.py, the MultiModalHasher class has a security and data integrity issue in its image hashing method. Currently, it serializes PIL.Image.Image...

Privacy-Preserving Prompt Personalization in Federated Learning for Multimodal Large Language Models

Prompt learning is a crucial technique for adapting pre-trained multimodal language models MLLMs to user tasks. Federated prompt personalization FPP is further developed to address data heterogeneity and local overfitting, however, it exposes personalized prompts - valuable intellectual assets - ...

Test-Time Immunization: a Universal Defense Framework against Jailbreaks for (Multimodal) Large Language Models

While multimodal large language models LLMs have attracted widespread attention due to their exceptional capabilities, they remain vulnerable to jailbreak attacks. Various defense methods are proposed to defend against jailbreak attacks, however, they are often tailored to specific types of...