Beyond Text: Multimodal Jailbreaking of Vision-Language and Audio Models through Perceptually Simple Transformations
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911228071247872 |
|---|---|
| author | Kumar, Divyanshu Jena, Shreyas Birur, Nitin Aravind Baswa, Tanay Agarwal, Sahil Harshangi, Prashanth |
| author_facet | Kumar, Divyanshu Jena, Shreyas Birur, Nitin Aravind Baswa, Tanay Agarwal, Sahil Harshangi, Prashanth |
| contents | Multimodal large language models (MLLMs) have achieved remarkable progress, yet remain critically vulnerable to adversarial attacks that exploit weaknesses in cross-modal processing. We present a systematic study of multimodal jailbreaks targeting both vision-language and audio-language models, showing that even simple perceptual transformations can reliably bypass state-of-the-art safety filters. Our evaluation spans 1,900 adversarial prompts across three high-risk safety categories harmful content, CBRN (Chemical, Biological, Radiological, Nuclear), and CSEM (Child Sexual Exploitation Material) tested against seven frontier models. We explore the effectiveness of attack techniques on MLLMs, including FigStep-Pro (visual keyword decomposition), Intelligent Masking (semantic obfuscation), and audio perturbations (Wave-Echo, Wave-Pitch, Wave-Speed). The results reveal severe vulnerabilities: models with almost perfect text-only safety (0\% ASR) suffer >75\% attack success under perceptually modified inputs, with FigStep-Pro achieving up to 89\% ASR in Llama-4 variants. Audio-based attacks further uncover provider-specific weaknesses, with even basic modality transfer yielding 25\% ASR for technical queries. These findings expose a critical gap between text-centric alignment and multimodal threats, demonstrating that current safeguards fail to generalize across cross-modal attacks. The accessibility of these attacks, which require minimal technical expertise, suggests that robust multimodal AI safety will require a paradigm shift toward broader semantic-level reasoning to mitigate possible risks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20223 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Text: Multimodal Jailbreaking of Vision-Language and Audio Models through Perceptually Simple Transformations Kumar, Divyanshu Jena, Shreyas Birur, Nitin Aravind Baswa, Tanay Agarwal, Sahil Harshangi, Prashanth Cryptography and Security Multimedia Multimodal large language models (MLLMs) have achieved remarkable progress, yet remain critically vulnerable to adversarial attacks that exploit weaknesses in cross-modal processing. We present a systematic study of multimodal jailbreaks targeting both vision-language and audio-language models, showing that even simple perceptual transformations can reliably bypass state-of-the-art safety filters. Our evaluation spans 1,900 adversarial prompts across three high-risk safety categories harmful content, CBRN (Chemical, Biological, Radiological, Nuclear), and CSEM (Child Sexual Exploitation Material) tested against seven frontier models. We explore the effectiveness of attack techniques on MLLMs, including FigStep-Pro (visual keyword decomposition), Intelligent Masking (semantic obfuscation), and audio perturbations (Wave-Echo, Wave-Pitch, Wave-Speed). The results reveal severe vulnerabilities: models with almost perfect text-only safety (0\% ASR) suffer >75\% attack success under perceptually modified inputs, with FigStep-Pro achieving up to 89\% ASR in Llama-4 variants. Audio-based attacks further uncover provider-specific weaknesses, with even basic modality transfer yielding 25\% ASR for technical queries. These findings expose a critical gap between text-centric alignment and multimodal threats, demonstrating that current safeguards fail to generalize across cross-modal attacks. The accessibility of these attacks, which require minimal technical expertise, suggests that robust multimodal AI safety will require a paradigm shift toward broader semantic-level reasoning to mitigate possible risks. |
| title | Beyond Text: Multimodal Jailbreaking of Vision-Language and Audio Models through Perceptually Simple Transformations |
| topic | Cryptography and Security Multimedia |
| url | https://arxiv.org/abs/2510.20223 |