Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Study

v1v2v3 (latest)

Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Study

21 May 2025

ArXiv (abs)PDF HTML HuggingFace (8 upvotes)Github (3★)

Papers citing "Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Study"

7 / 7 papers shown

Title
Jailbreaking on Text-to-Video Models via Scene Splitting Strategy Wonjun Lee Haon Park Doehyeon Lee Bumsub Ham Suhyun Kim 8 0 0 26 Sep 2025
Better Safe Than Sorry? Overreaction Problem of Vision Language Models in Visual Emergency Recognition Dasol Choi Seunghyun Lee Youngsook Song 139 2 0 21 May 2025
Output Constraints as Attack Surface: Exploiting Structured Generation to Bypass LLM Safety Mechanisms Shuoming Zhang Jiacheng Zhao Ruiyuan Xu Xiaobing Feng Huimin Cui AAML 127 5 0 31 Mar 2025
Exploiting Prefix-Tree in Structured Output Interfaces for Enhancing Jailbreak Attacking Yanzeng Li Yunfan Xiong Jialun Zhong Jinchao Zhang Jie Zhou Lei Zou 81 3 0 20 Feb 2025
SafeDialBench: A Fine-Grained Safety Benchmark for Large Language Models in Multi-Turn Dialogues with Diverse Jailbreak Attacks Hongye Cao Yanming Wang Sijia Jing Ziyue Peng Zhixin Bai ... Yang Gao Fanyu Meng Xi Yang Chao Deng Junlan Feng AAML 224 4 0 16 Feb 2025
Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack M. Russinovich Ahmed Salem Ronen Eldan 227 140 0 02 Apr 2024
FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts Yichen Gong Delong Ran Jinyuan Liu Conglei Wang Tianshuo Cong Anyu Wang Sisi Duan Xiaoyun Wang MLLM 372 222 0 09 Nov 2023