cs.CRNov 18, 2026

What Survives the Shift from Text to Pixels? Conditional Protection against Prompt Injection in Image-Rendered Documents

Authors: Haoyang Liu, Yinong Xu, Zhe Lang, Zhenyu Guan, Qianyun Zhang

Organizations: School of Cyber Science and Technology, Beihang University

Abstract

Multimodal language models increasingly answer questions over external documents, exposing them to indirect prompt injection: commands planted in untrusted content to hijack the response. Rendering untrusted content as page images lowers attack success on average, leaving open which injected operations survive rendering, at what utility cost, and whether failed image attacks reflect unread commands or commands read but not followed. We pair text and page-image versions of 191 English question-answering sources across six models, measuring marker appending, answer formatting, and answer replacement alongside clean Question-Answering utility. Rendering reduces append and format execution for every model, yet Qwen3-VL-8B still executes image-borne replacement at 77.96%. Protection also depends on target type, rendering settings, wording, and repetition, and a search over eight fixed attack candidates largely closes the image deficit in appending and formatting for Qwen and Gemini, though not for GPT-5-nano. Utility ratios on original questions range from 93.31% to 101.61%, but Qwen falls to 61.99% when answers combine two evidence spans. Illegibility cannot explain all failed attacks: models often transcribe and identify image-borne commands they do not execute, and trusted user authorization raises execution on unchanged pages. Inside Qwen, patching early attack-region states induces all three operations, more weakly in images. Attention blocking suggests why: a text command acts through its own positions, whereas visual encoding spreads an image command and its target across the instruction page. Rendering thus offers conditional protection, motivating evaluations that report security by model, operation, and attack budget alongside utility.

Figures & tables

Explore similar work

CardsList
  1. Render Before Reading: Visual Rendering as a Prompt Injection Defense

    Sep 28, 2026Jie Zhang, Andrei Baroian, Jan N. van Rijn +2Multimodal Large Language ModelsModality Gap in VLMs

  2. Repeat-After-Me: Black-Box Adaptive Visual Prompt Injection

    Sep 3, 2026Sizhe Chen, Yu-Lin Tsai, Ivan Evtimov +4Visual Prompt Injection AttacksPrompt Injection Attacks

  3. Image-based Prompt Injection: Hijacking Multimodal LLMs through Visually Embedded Adversarial Instructions

    Mar 4, 2026Neha Nagaraja, Lan Zhang, Zhilong Wang +2Visual Prompt Injection AttacksPrompt Injection Attacks