2024/03/19 by Anjun Hu, Jindong Gu, Hu, Anjun +7 · 3 citations
Computer Science · #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences
paper · pdf · doi:10.48550/arxiv.2403.12693
openalex publication_date 2024/03/19 · openalex created_date 2024/03/21 · openalex updated_date 2026/07/28
Foundation models pre-trained on web-scale vision-language data, such as CLIP, are widely used as cornerstones of powerful machine learning systems. While pre-training offers clear advantages for downstream learning, it also endows downstream models with shared adversarial vulnerabilities that can be easily identified through the open-sourced foundation model. In this work, we expose such vulnerabilities in CLIP's downstream models and show that foundation models can serve as a basis for attacking their downstream systems. In particular, we propose a simple yet effective adversarial attack strategy termed Patch Representation Misalignment (PRM). Solely based on open-sourced CLIP vision encoders, this method produces adversaries that simultaneously fool more than 20 downstream models spanning 4 common vision-language tasks (semantic segmentation, object detection, image captioning and visual question-answering). Our findings highlight the concerning safety risks introduced by the extensive usage of public foundational models in the development of downstream systems, calling for extra caution in these scenarios.