A Systematic Study of Cross-Modal Typographic Attacks on Audio-Visual Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Tianle, Ghadiyaram, Deepti
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908937383575552
author Chen, Tianle
Ghadiyaram, Deepti
author_facet Chen, Tianle
Ghadiyaram, Deepti
contents As audio-visual multi-modal large language models (MLLMs) are increasingly deployed in safety-critical applications, understanding their vulnerabilities is crucial. To this end, we introduce Multi-Modal Typography, a systematic study examining how typographic attacks across multiple modalities adversely influence MLLMs. While prior work focuses narrowly on unimodal attacks, we expose the cross-modal fragility of MLLMs. We analyze the interactions between audio, visual, and text perturbations and reveal that coordinated multi-modal attack creates a significantly more potent threat than single-modality attacks (attack success rate = $83.43\%$ vs $34.93\%$).Our findings across multiple frontier MLLMs, tasks, and common-sense reasoning and content moderation benchmarks establishes multi-modal typography as a critical and underexplored attack strategy in multi-modal reasoning. Code and data will be publicly available.
format Preprint
id arxiv_https___arxiv_org_abs_2604_03995
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Systematic Study of Cross-Modal Typographic Attacks on Audio-Visual Reasoning
Chen, Tianle
Ghadiyaram, Deepti
Computer Vision and Pattern Recognition
Sound
As audio-visual multi-modal large language models (MLLMs) are increasingly deployed in safety-critical applications, understanding their vulnerabilities is crucial. To this end, we introduce Multi-Modal Typography, a systematic study examining how typographic attacks across multiple modalities adversely influence MLLMs. While prior work focuses narrowly on unimodal attacks, we expose the cross-modal fragility of MLLMs. We analyze the interactions between audio, visual, and text perturbations and reveal that coordinated multi-modal attack creates a significantly more potent threat than single-modality attacks (attack success rate = $83.43\%$ vs $34.93\%$).Our findings across multiple frontier MLLMs, tasks, and common-sense reasoning and content moderation benchmarks establishes multi-modal typography as a critical and underexplored attack strategy in multi-modal reasoning. Code and data will be publicly available.
title A Systematic Study of Cross-Modal Typographic Attacks on Audio-Visual Reasoning
topic Computer Vision and Pattern Recognition
Sound
url https://arxiv.org/abs/2604.03995