paperApril 2024Unreviewed

Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?

Shuo Chen, Zhen Han, Bailan He, Zifeng Ding, Wenqian Yu, Philip H. S. Torr, Volker Tresp, Jindong Gu

arXiv.org

Abstract

Various jailbreak attacks have been proposed to red-team Large Language Models (LLMs) and revealed the vulnerable safeguards of LLMs. Besides, some methods are not limited to the textual modality and extend the jailbreak attack to Multimodal Large Language Models (MLLMs) by perturbing the visual input. However, the absence of a universal evaluation benchmark complicates the performance reproduction and fair comparison. Besides, there is a lack of comprehensive evaluation of closed-source state-o

Framework mappings

OWASP Top 10 for LLM Applications

LLM01Prompt Injection

MITRE ATLAS

AML.T0054LLM Jailbreak

NIST AI Risk Management Framework

MEASUREMeasure

Suggested from the entry's categories.

Cite

@misc{chen2024red,
  title = {{Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?}},
  author = {Shuo Chen and Zhen Han and Bailan He and Zifeng Ding and Wenqian Yu and Philip H. S. Torr and Volker Tresp and Jindong Gu},
  year = {2024},
  month = apr,
  eprint = {2404.03411},
  archivePrefix = {arXiv},
  doi = {10.48550/arXiv.2404.03411},
  url = {https://www.semanticscholar.org/paper/526c88e301af88080ae4ecc3b65c9fc1f8f383f8}
}