# adversarial-images
標記為「adversarial-images」的 5 篇文章
Case Study: GPT-4 Vision Jailbreak Attacks
Analysis of visual jailbreak techniques targeting GPT-4V's multimodal capabilities, including typography attacks, adversarial images, and cross-modal prompt injection.
case-studiesgpt-4vmultimodalvisionjailbreakadversarial-images
Cross-Modal Embedding Attacks
Exploitation of shared embedding spaces across modalities: CLIP adversarial images, typographic attacks, VLM injection, and dimensionality reduction attacks.
CLIPmultimodalcross-modalvision-languageadversarial-images
Case Study: GPT-4 Vision 越獄 攻擊s
Analysis of visual jailbreak techniques targeting GPT-4V's multimodal capabilities, including typography attacks, adversarial images, and cross-modal prompt injection.
case-studiesgpt-4vmultimodalvisionjailbreakadversarial-images
跨模態 Embedding 攻擊
跨模態之共享 embedding 空間利用：CLIP 對抗影像、印刷式攻擊、VLM 注入與維度縮減攻擊。
CLIPmultimodalcross-modalvision-languageadversarial-images
視覺-語言模型
視覺-語言模型（VLM）的安全評估——涵蓋 VLM 架構、圖片注入技術、OCR 與字型攻擊、對抗性圖片生成與 VLM 特定越獄。
vlmvisionimage-injectionocradversarial-imagesmultimodal