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

arXiv: https://arxiv.org/abs/2609.26733 · HTML: https://mlsanity.com/paper/2609.26733/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 11 of its 26 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM jailbreaks and adversarial robustness](https://mlsanity.com/mllm/mllm-jailbreaks-and-adversarial-robustness/)

## Builds on (most cited first, 11 of 11)

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Generalisation in humans and deep neural networks](https://mlsanity.com/paper/1808.08750/) (2018)
- [On the Robustness of Vision Transformers to Adversarial Examples](https://mlsanity.com/paper/2104.02610/) (2021)
- [Testing of Detection Tools for AI-Generated Text](https://mlsanity.com/paper/2306.15666/) (2023)
- [How Large Language Models are Transforming Machine-Paraphrased Plagiarism](https://mlsanity.com/paper/2210.03568/) (2022)
- [InternSpatial: A Comprehensive Dataset for Spatial Reasoning in Vision-Language Models](https://mlsanity.com/paper/2506.18385/) (2025)
- [Game of Tones: Faculty detection of GPT-4 generated content in university assessments](https://mlsanity.com/paper/2305.18081/) (2023)
- [Are Your Explanations Reliable? Investigating the Stability of LIME in Explaining Text Classifiers by Marrying XAI and Adversarial Attack](https://mlsanity.com/paper/2305.12351/) (2023)
- [Impeding LLM-assisted Cheating in Introductory Programming Assignments via Adversarial Perturbation](https://mlsanity.com/paper/2410.09318/) (2024)
- [Can AI Recognize Its Own Reflection? Self-Detection Performance of LLMs in Computing Education](https://mlsanity.com/paper/2512.23587/) (2025)
- [Inappropriate Benefits and Identification of ChatGPT Misuse in Programming Tests: A Controlled Experiment](https://mlsanity.com/paper/2309.16697/) (2023)

## Abstract (arXiv)

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.