Multi-Paradigm Collaborative Adversarial Attack Against Multi-Modal Large Language Models
This research addresses the critical vulnerability of Multi-Modal Large Language Models to adversarial attacks, which is a significant concern for the security and reliability of AI systems.
This paper introduces a novel adversarial attack framework, MPCAttack, designed to enhance the transferability of adversarial examples against Multi-Modal Large Language Models (MLLMs). By aggregating semantic representations from both visual images and language texts and employing a Multi-Paradigm Collaborative Optimisation strategy, MPCAttack consistently outperforms state-of-the-art methods in both targeted and untargeted attacks on various MLLMs.
The rapid progress of Multi-Modal Large Language Models (MLLMs) has significantly advanced downstream applications. However, this progress also exposes serious transferable adversarial vulnerabilities. In general, existing adversarial attacks against MLLMs typically rely on surrogate models trained within a single learning paradigm and perform independent optimisation in their respective feature spaces. This straightforward setting naturally restricts the richness of feature representations, delivering limits on the search space and thus impeding the diversity of adversarial perturbations. To address this, we propose a novel Multi-Paradigm Collaborative Attack (MPCAttack) framework to boost the transferability of adversarial examples against MLLMs. In principle, MPCAttack aggregates semantic representations, from both visual images and language texts, to facilitate joint adversarial optimisation on the aggregated features through a Multi-Paradigm Collaborative Optimisation (MPCO) strategy. By performing contrastive matching on multi-paradigm features, MPCO adaptively balances the importance of different paradigm representations and guides the global perturbation optimisation, effectively alleviating the representation bias. Extensive experimental results on multiple benchmarks demonstrate the superiority of MPCAttack, indicating that our solution consistently outperforms state-of-the-art methods in both targeted and untargeted attacks on open-source and closed-source MLLMs. The code is released at https://github.com/LiYuanBoJNU/MPCAttack.