CAVALRY-V: A Large-Scale Generator Framework for Adversarial Attacks on Video MLLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Jiaming, Hu, Rui, Guo, Qing, Lim, Wei Yang Bryan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908429463846912
author Zhang, Jiaming
Hu, Rui
Guo, Qing
Lim, Wei Yang Bryan
author_facet Zhang, Jiaming
Hu, Rui
Guo, Qing
Lim, Wei Yang Bryan
contents Video Multimodal Large Language Models (V-MLLMs) have shown impressive capabilities in temporal reasoning and cross-modal understanding, yet their vulnerability to adversarial attacks remains underexplored due to unique challenges: complex cross-modal reasoning mechanisms, temporal dependencies, and computational constraints. We present CAVALRY-V (Cross-modal Language-Vision Adversarial Yielding for Videos), a novel framework that directly targets the critical interface between visual perception and language generation in V-MLLMs. Our approach introduces two key innovations: (1) a dual-objective semantic-visual loss function that simultaneously disrupts the model's text generation logits and visual representations to undermine cross-modal integration, and (2) a computationally efficient two-stage generator framework that combines large-scale pre-training for cross-model transferability with specialized fine-tuning for spatiotemporal coherence. Empirical evaluation on comprehensive video understanding benchmarks demonstrates that CAVALRY-V significantly outperforms existing attack methods, achieving 22.8% average improvement over the best baseline attacks on both commercial systems (GPT-4.1, Gemini 2.0) and open-source models (QwenVL-2.5, InternVL-2.5, Llava-Video, Aria, MiniCPM-o-2.6). Our framework achieves flexibility through implicit temporal coherence modeling rather than explicit regularization, enabling significant performance improvements even on image understanding (34.4% average gain). This capability demonstrates CAVALRY-V's potential as a foundational approach for adversarial research across multimodal systems.
format Preprint
id arxiv_https___arxiv_org_abs_2507_00817
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CAVALRY-V: A Large-Scale Generator Framework for Adversarial Attacks on Video MLLMs
Zhang, Jiaming
Hu, Rui
Guo, Qing
Lim, Wei Yang Bryan
Computer Vision and Pattern Recognition
Artificial Intelligence
Video Multimodal Large Language Models (V-MLLMs) have shown impressive capabilities in temporal reasoning and cross-modal understanding, yet their vulnerability to adversarial attacks remains underexplored due to unique challenges: complex cross-modal reasoning mechanisms, temporal dependencies, and computational constraints. We present CAVALRY-V (Cross-modal Language-Vision Adversarial Yielding for Videos), a novel framework that directly targets the critical interface between visual perception and language generation in V-MLLMs. Our approach introduces two key innovations: (1) a dual-objective semantic-visual loss function that simultaneously disrupts the model's text generation logits and visual representations to undermine cross-modal integration, and (2) a computationally efficient two-stage generator framework that combines large-scale pre-training for cross-model transferability with specialized fine-tuning for spatiotemporal coherence. Empirical evaluation on comprehensive video understanding benchmarks demonstrates that CAVALRY-V significantly outperforms existing attack methods, achieving 22.8% average improvement over the best baseline attacks on both commercial systems (GPT-4.1, Gemini 2.0) and open-source models (QwenVL-2.5, InternVL-2.5, Llava-Video, Aria, MiniCPM-o-2.6). Our framework achieves flexibility through implicit temporal coherence modeling rather than explicit regularization, enabling significant performance improvements even on image understanding (34.4% average gain). This capability demonstrates CAVALRY-V's potential as a foundational approach for adversarial research across multimodal systems.
title CAVALRY-V: A Large-Scale Generator Framework for Adversarial Attacks on Video MLLMs
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2507.00817