Evaluating the Semantic-to-Geometric Gap in Adversarial Defenses Against Vision-Language Model-Based Plagiarism

2026-09-22

AI Summary

研究通过评估对抗性图像变换来解决视觉-语言模型在学术抄袭中的应用问题，发现这些模型对对抗扰动敏感，建议重新设计评估方法。

Abstract

The rapidly advancing capabilities of vision-language models (VLMs) present a systemic challenge to academic integrity. VLMs now allow students to bypass meaningful engagement by capturing and submitting graphical problems as singular images, a practice we define as trivial plagiarism. To provide educators with actionable data on VLM limitations, we investigate the efficacy of heuristic adversarial image transformations designed to degrade model performance while remaining human-interpretable. Through a two-phase evaluation of introductory assessments, we manually assess baseline VLM performance on circuit diagrams, followed by an automated large-scale evaluation of topological structures (logic gates) and coordinate geometry (Karnaugh maps). We find that while highly capable VLMs can exhibit appreciable robustness, all models suffer vulnerability to adversarial perturbations. We conclude that while visual perturbations act as a viable near-term stopgap, long-term assessment security requires educators to reapproach assessment design given continually increasing VLM performance.

Problem

• vision-language models
• academic integrity
• trivial plagiarism
• adversarial defenses
• heuristic adversarial image transformations

Innovation

• adversarial image transformations
• vision-language models
• trivial plagiarism
• assessment security
• robustness