On the Robustness of Large Multimodal Models Against Image Adversarial Attacks
This addresses a critical security vulnerability for users of multimodal AI systems, though it is incremental as it builds on existing adversarial attack research.
The study investigated the robustness of Large Multimodal Models (LMMs) against image adversarial attacks, finding they are generally not robust, but context from prompts like questions can mitigate effects, with only an 8.10% performance drop on ScienceQA compared to 99.73% for visual models.
Recent advances in instruction tuning have led to the development of State-of-the-Art Large Multimodal Models (LMMs). Given the novelty of these models, the impact of visual adversarial attacks on LMMs has not been thoroughly examined. We conduct a comprehensive study of the robustness of various LMMs against different adversarial attacks, evaluated across tasks including image classification, image captioning, and Visual Question Answer (VQA). We find that in general LMMs are not robust to visual adversarial inputs. However, our findings suggest that context provided to the model via prompts, such as questions in a QA pair helps to mitigate the effects of visual adversarial inputs. Notably, the LMMs evaluated demonstrated remarkable resilience to such attacks on the ScienceQA task with only an 8.10% drop in performance compared to their visual counterparts which dropped 99.73%. We also propose a new approach to real-world image classification which we term query decomposition. By incorporating existence queries into our input prompt we observe diminished attack effectiveness and improvements in image classification accuracy. This research highlights a previously under-explored facet of LMM robustness and sets the stage for future work aimed at strengthening the resilience of multimodal systems in adversarial environments.