Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models

arXiv:2404.12916v1 Announce Type: new
Abstract: Vision-Large-Language-models(VLMs) have great application prospects in autonomous driving. Despite the ability of VLMs to comprehend and make decisions in complex scenarios, their integration into safety-critical autonomous driving systems poses serious security risks. In this paper, we propose BadVLMDriver, the first backdoor attack against VLMs for autonomous driving that can be launched in practice using physical objects. Unlike existing backdoor attacks against VLMs that rely on digital modifications, BadVLMDriver uses common physical items, such as a …

application arxiv attack autonomous autonomous driving backdoor backdoor attack can critical cs.cr driving great integration language language models large large-language-models physical risks safety safety-critical security security risks serious serious security systems