mlsanityData to 7 Oct 2026

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

arXiv 2604.03995 ↗First posted 5 Apr 2026cs.CV cs.SDcited by 0

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 17 of its 37 references that are in the map. First posted on arXiv 5 Apr 2026.

Builds on

17 of 37 references.

References · 17

  1. 2023PandaGPT: One Model To Instruction-Follow Them Allcited by 330
  2. 2025WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMscited by 147
  3. 2018Adversarial Attacks Against Automatic Speech Recognition Systems via Psychoacoustic Hidingcited by 83
  4. 2024AVHBench: A Cross-Modal Hallucination Benchmark for Audio-Visual Large Language Modelscited by 50
  5. 2025Voxtralcited by 41
  6. 2025MLLMs are Deeply Affected by Modality Biascited by 38
  7. 2024Vision-LLMs Can Fool Themselves with Self-Generated Typographic Attackscited by 36
  8. 2024Safeguarding Vision-Language Models Against Patched Visual Prompt Injectorscited by 26
  9. 2025Multilingual and Multi-Accent Jailbreaking of Audio LLMscited by 25
  10. 2024Empirical Analysis of Large Vision-Language Models against Goal Hijacking via Visual Prompt Injectioncited by 17
  11. 2025Jailbreak-AudioBench: In-Depth Evaluation and Analysis of Jailbreak Threats for Large Audio Language Modelscited by 17
  12. 2025Exploring Typographic Visual Prompts Injection Threats in Cross-Modality Generation Modelscited by 12
5 more
  1. 2025Dyslexify: A Mechanistic Defense Against Typographic Attacks in CLIPcited by 8
  2. 2025Some Modalities are More Equal Than Others: Decoding and Architecting Multimodal Integration in MLLMscited by 7
  3. 2026Now You Hear Me: Audio Narrative Attacks Against Large Audio-Language Modelscited by 5
  4. 2026Physical Prompt Injection Attacks on Large Vision-Language Modelscited by 3
  5. 2024SLANT: Spurious Logo ANalysis Toolkitcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Unveiling Typographic Deceptions: Insights of the Typographic Vulnerability in Large Vision-Language Model
  2. 2025Investigating Vulnerabilities and Defenses Against Audio-Visual Attacks: A Comprehensive Survey Emphasizing Multimodal Models
  3. 2025SCAM: A Real-World Typographic Robustness Evaluation for Multimodal Foundation Models
  4. 2026CSR-Bench: A Benchmark for Evaluating the Cross-modal Safety and Reliability of MLLMs
  5. 2025VRSA: Jailbreaking Multimodal Large Language Models through Visual Reasoning Sequential Attack
  6. 2026Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Survey
  7. 2025PolyJailbreak: Cross-Modal Jailbreaking Attacks on Black-Box Multimodal LLMs
  8. 2025Omni-SafetyBench: A Benchmark for Safety Evaluation of Audio-Visual Large Language Models