EO-VLM: VLM-Guided Energy Overload Attacks on Vision Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915236840210432 |
|---|---|
| author | Seo, Minjae You, Myoungsung Lee, Junhee Kim, Jaehan Heo, Hwanjo Oh, Jintae Kim, Jinwoo |
| author_facet | Seo, Minjae You, Myoungsung Lee, Junhee Kim, Jaehan Heo, Hwanjo Oh, Jintae Kim, Jinwoo |
| contents | Vision models are increasingly deployed in critical applications such as autonomous driving and CCTV monitoring, yet they remain susceptible to resource-consuming attacks. In this paper, we introduce a novel energy-overloading attack that leverages vision language model (VLM) prompts to generate adversarial images targeting vision models. These images, though imperceptible to the human eye, significantly increase GPU energy consumption across various vision models, threatening the availability of these systems. Our framework, EO-VLM (Energy Overload via VLM), is model-agnostic, meaning it is not limited by the architecture or type of the target vision model. By exploiting the lack of safety filters in VLMs like DALL-E 3, we create adversarial noise images without requiring prior knowledge or internal structure of the target vision models. Our experiments demonstrate up to a 50% increase in energy consumption, revealing a critical vulnerability in current vision models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_08205 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | EO-VLM: VLM-Guided Energy Overload Attacks on Vision Models Seo, Minjae You, Myoungsung Lee, Junhee Kim, Jaehan Heo, Hwanjo Oh, Jintae Kim, Jinwoo Computer Vision and Pattern Recognition Cryptography and Security Vision models are increasingly deployed in critical applications such as autonomous driving and CCTV monitoring, yet they remain susceptible to resource-consuming attacks. In this paper, we introduce a novel energy-overloading attack that leverages vision language model (VLM) prompts to generate adversarial images targeting vision models. These images, though imperceptible to the human eye, significantly increase GPU energy consumption across various vision models, threatening the availability of these systems. Our framework, EO-VLM (Energy Overload via VLM), is model-agnostic, meaning it is not limited by the architecture or type of the target vision model. By exploiting the lack of safety filters in VLMs like DALL-E 3, we create adversarial noise images without requiring prior knowledge or internal structure of the target vision models. Our experiments demonstrate up to a 50% increase in energy consumption, revealing a critical vulnerability in current vision models. |
| title | EO-VLM: VLM-Guided Energy Overload Attacks on Vision Models |
| topic | Computer Vision and Pattern Recognition Cryptography and Security |
| url | https://arxiv.org/abs/2504.08205 |