相关论文: Safety Geometry Collapse in Multimodal LLMs and Ad…

Instruction Fine-Tuning (IFT) has been widely adopted as an effective post-training strategy to enhance various abilities of Large Language Models (LLMs). However, prior studies have shown that IFT can significantly compromise LLMs' safety,…

Safety-aligned language models refuse harmful requests through learned refusal behaviors encoded in their internal representations. Recent activation-based jailbreaking methods circumvent these safety mechanisms by applying orthogonal…

Following road safety norms is non-negotiable not only for humans but also for the AI systems that govern autonomous vehicles. In this work, we evaluate how well multi-modal large language models (LLMs) understand road safety concepts,…

Multimodal Large Language Models (MLLMs) extend text-only LLMs with visual reasoning, but also introduce new safety failure modes under visually grounded instructions. We study comic-template jailbreaks that embed harmful goals inside…

Training multimodal large language models has long been limited by the scarcity of high-quality paired multimodal data. Recent studies show that the shared representation space of pretrained multimodal contrastive models can serve as a…

Multimodal large language models (MLLMs) exhibit remarkable capabilities but remain susceptible to jailbreak attacks exploiting cross-modal vulnerabilities. In this work, we introduce a novel method that leverages sequential comic-style…

This paper presents a systematic evaluation of Large Language Models' (LLMs) behavior on long-tail distributed (encrypted) texts and their safety implications. We introduce a two-dimensional framework for assessing LLM safety: (1)…

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…

Large Language Models (LLMs), primarily trained on text-based datasets, exhibit exceptional proficiencies in understanding and executing complex linguistic instructions via text outputs. However, they falter when requests to generate…

The rapid development of Multimodal Large Reasoning Models (MLRMs) has demonstrated broad application potential, yet their safety and reliability remain critical concerns that require systematic exploration. To address this gap, we conduct…

We explore Multimodal Large Language Models (MLLMs), which integrate LLMs like GPT-4 to handle multimodal data, including text, images, audio, and more. MLLMs demonstrate capabilities such as generating image captions and answering…

Refusal mechanisms in large language models (LLMs) are essential for ensuring safety. Recent research has revealed that refusal behavior can be mediated by a single direction in activation space, enabling targeted interventions to bypass…

Large Multimodal Models (LMMs) have demonstrated impressive capabilities in video reasoning via Chain-of-Thought (CoT). However, the robustness of their reasoning chains remains questionable. In this paper, we identify a critical failure…

End-to-end Large Speech Language Models (LSLMs) have demonstrated impressive conversational generation abilities, yet consistently fall short of traditional pipeline systems on semantic understanding benchmarks. In this work, we reveal…

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks, yet they often refuse to answer legitimate queries--a phenomenon known as overrefusal. Overrefusal typically stems from over-conservative…

Open-weight LLMs can be modified at inference time with simple activation edits, which raises a practical question for safety: do common safety interventions like refusal training or metatag training survive such edits? We study model…

Multi-Modal Language Models (MLLMs) have transformed artificial intelligence by combining visual and text data, making applications like image captioning, visual question answering, and multi-modal content creation possible. This ability to…

Large language models (LLMs) can translate natural language into optimization code, but silent failures pose a critical risk: code that executes and returns solver-feasible solutions may encode semantically incorrect formulations -- a…

Multi-modal Large Language Models (MLLMs) frequently face challenges from concept drift when dealing with real-world streaming data, wherein distributions change unpredictably. This mainly includes gradual drift due to long-tailed data and…

Traditional approaches to safety event analysis in autonomous systems have relied on complex machine learning models and extensive datasets for high accuracy and reliability. However, the advent of Multimodal Large Language Models (MLLMs)…