发表机构
Nanjing University; Meituan; Yale University; Institute of Automation, Chinese Academy of Sciences（南京大学; 美团; 耶鲁大学; 中国科学院自动化研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对多模态大语言模型的安全评估，提出细粒度基准MME-Safety，采用四维标注和分层评估框架，在17个模型上揭示跨模态风险及思维链推理的安全隐患。
AI 中文摘要
尽管多模态大语言模型（MLLMs）展现出显著进步，但其跨模态能力引入了复杂的脆弱性，这些脆弱性容易绕过单模态过滤器。现有基准缺乏细粒度的意图相关标注，并依赖单一维度指标，阻碍了全面的鲁棒性评估。为解决这一问题，我们提出了MME-Safety，一个经过严格验证的基准，具有独特的四维标注模式，对风险场景、危害严重性和模态特定隐蔽级别进行分类。此外，我们引入了一个分层评估框架，以评估基础响应可靠性、实际风险暴露以及防御行为的结构完整性。对17个最先进MLLMs的广泛零样本评估提供了当前多模态系统的全面安全概况。我们的分析系统性地研究了跨模态输入配置，并揭示了与思维链（CoT）推理相关的安全隐患。这些多方面的发现强调了在多模态领域进行稳健的、具有推理意识的安全对齐的紧迫需求。
英文摘要
While Multimodal Large Language Models (MLLMs) show remarkable advancements, their cross-modal capabilities introduce complex vulnerabilities that easily bypass unimodal filters. Existing benchmarks lack fine-grained intent-related annotations and rely on unidimensional metrics, hindering comprehensive robustness evaluation. To address this, we propose MME-Safety, a rigorously verified benchmark featuring a unique four-dimensional annotation schema that categorizes risk scenarios, harm severity, and modality-specific stealth levels. Furthermore, we introduce a hierarchical evaluation framework to assess fundamental response reliability, actual risk exposure, and the structural integrity of defensive behaviors. Extensive zero-shot evaluations across 17 state-of-the-art MLLMs provide a comprehensive safety profile of current multimodal systems. Our analysis systematically investigates cross-modal input configurations and uncovers safety implications associated with Chain-of-Thought (CoT) reasoning. These multifaceted findings underscore the urgent need for robust, reasoning-aware safety alignment in the multimodal landscape.
