Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

OCR detection

OCR detection, also known as optical character text detection, is the computer vision process of identifying and localizing the presence of textual content within an image or video. Serving as the foundational first stage of an optical character recognition pipeline before text recognition and transcription occur, OCR detection determines the precise spatial boundaries, such as bounding boxes or polygon coordinates, around printed, handwritten, or typographic text. This capability enables automated systems and multimodal artificial intelligence models to isolate embedded text from complex visual backgrounds, allowing the detected typographic elements to be accurately read, indexed, or analyzed for downstream processing and content filtering.

1 item

FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts

FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts

Yichen Gong, Delong Ran, Jinyuan Liu, Conglei Wang, Tianshuo Cong, Anyu Wang, Sisi Duan, Xiaoyun Wang

Why you should read this

Presents FigStep, a lightweight black-box jailbreak method that bypasses vision-language model safeguards by converting forbidden text into typographic images, exposing critical cross-modal alignment gaps across both open-source and proprietary systems.

Large Vision-Language Models (LVLMs) signify a groundbreaking paradigm shift within the Artificial Intelligence (AI) community, extending beyond the capabilities of Large Language Models (LLMs) by assimilating additional modalities (e.g., images). Despite this advancement, the safety of LVLMs remains adequately underexplored, with a potential overreliance on the safety assurances purportedly by their underlying LLMs. In this paper, we propose FigStep, a straightforward yet effective black-box jailbreak algorithm against LVLMs. Instead of feeding textual harmful instructions directly, FigStep converts the prohibited content into images through typography to bypass the safety alignment. The experimental results indicate that FigStep can achieve an average attack success rate of 82.50% on six promising open-source LVLMs. Not merely to demonstrate the efficacy of FigStep, we conduct comprehensive ablation studies and analyze the distribution of the semantic embeddings to uncover that the reason behind the success of FigStep is the deficiency of safety alignment for visual embeddings. Moreover, we compare FigStep with five text-only jailbreaks and four image-based jailbreaks to demonstrate the superiority of FigStep, i.e., negligible attack costs and better attack performance. Above all, our work reveals that current LVLMs are vulnerable to jailbreak attacks, which highlights the necessity of novel cross-modality safety alignment techniques.

Added

2026-09-26