AIRRI: Towards a Human-Readable, Machine-Resistant Reading Interface

📄 arXiv: 2610.18724v1 📥 PDF

作者: Emory Salaun, Syed Ayon, Kaleel Mahmood, Shaun Wallace

分类: cs.CV, cs.HC, cs.LG

发布日期: 2026-10-13

备注: 12 pages, 2 figures, 6 tables

💡 一句话要点

提出AIRRI以解决教育评估中的机器文本提取问题

🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 教育评估 图像到文本 对抗攻击 机器学习 文本提取 人工智能 抗性设计

📋 核心要点

  1. 现有的图像到文本提取技术使得教育评估面临挑战，作业材料易于被自动化处理，降低了评估的有效性。
  2. AIRRI通过图像变换和对抗攻击技术生成抗机器提取的教育内容，旨在提高作业的人工阅读性和抗性。
  3. 实验结果显示，AIRRI显著提高了字符和单词错误率，EasyOCR和Tesseract对抗扰动的敏感性高于GOT-OCR2模型。

📝 摘要（中文）

随着图像到文本提取系统和大型语言模型的快速发展，教育评估面临新的挑战，使得作业材料易于转录和解决。为应对这一问题，本文提出了AIRRI（人工智能抗性阅读接口），该框架旨在生成抗机器提取的教育内容，并提供高等教育作业的数据集用于评估。AIRRI采用了两类抗机器提取技术，包括图像变换和基于决策的对抗攻击。实验结果表明，AIRRI显著降低了自动文本提取的性能，增加了字符和单词错误率，表明其在AI辅助作业完成中的有效性。

🔬 方法详解

问题定义：本文旨在解决教育评估中作业材料被自动化文本提取系统轻易处理的问题，现有方法在防止机器提取方面存在不足。

核心思路：AIRRI的核心思路是通过图像变换和对抗攻击技术生成抗机器提取的文本内容，从而增加AI完成作业的难度。

技术框架：AIRRI框架包括两个主要模块：图像变换（如字符间距、单词间距、图像噪声等）和决策基对抗攻击（如RayS、SurFree和ADBA）。

关键创新：AIRRI的创新在于结合了多种图像变换和对抗攻击技术，显著提高了文本提取的难度，与现有方法相比，提供了更强的抗性。

关键设计：在设计中，AIRRI对图像进行了多种变换，并在实验中评估了不同提取系统的性能，特别是ADBA攻击在降低文本提取性能方面表现突出。

🖼️ 关键图片

fig_0
fig_1

📊 实验亮点

实验结果显示，AIRRI显著降低了自动文本提取的性能，EasyOCR的字符错误率从6.9%增加到55.9%，Tesseract的字符错误率从2.3%增加到100.0%。ADBA攻击在降低文本提取性能方面效果最佳，表明AIRRI在提高作业抗性方面的有效性。

🎯 应用场景

AIRRI的研究成果在教育领域具有广泛的应用潜力，尤其是在考试和作业评估中，可以有效防止学生利用自动化工具作弊，提升评估的公正性和有效性。未来，AIRRI的技术可以扩展到其他需要保护内容的领域，如版权保护和信息安全。

📄 摘要（原文）

The rapid advancement of image-to-text extraction systems and large language models (LLMs) poses new challenges for educational assessment by making assignment materials easy to transcribe and solve with minimal human effort. To address this problem, we present AIRRI (Artificial Intelligence Resilient Reading Interface), a framework for generating machine-resistant educational content, a dataset of higher-education assignments for evaluation, and an initial evaluation of the human readability of AIRRI machine-resistant text designs. AIRRI employs two categories of machine-resistance techniques. First, AIRRI applies image transformations including character spacing, word spacing, line spacing, character opacity variation, stripe overlays, and image noise. Second, AIRRI applies decision-based adversarial attacks: RayS, SurFree, and ADBA. We evaluate AIRRI against three image-to-text systems: Tesseract, EasyOCR, and the transformer-based GOT-OCR2 model. Our results show that AIRRI substantially degrades automated text extraction performance. AIRRI increased character error rates for EasyOCR (6.9% to 55.9%), Tesseract (2.3% to 100.0%), and GOT-OCR2 (30.7% to 48.1%). Word error rates increased for EasyOCR (24.9% to 96.7%) and Tesseract (10.7% to 100.0%). Among the evaluated decision-based attacks, ADBA consistently produced the largest degradation in image-to-text performance, while RayS and SurFree produced more moderate reductions. EasyOCR and Tesseract were generally more susceptible to both image-based and adversarial perturbations than the transformer-based GOT-OCR2 model. Our findings suggest that AIRRI can significantly reduce the reliability of image-to-text-to-LLM workflows, increasing the effort required for AI-assisted assignment completion.