Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.07947 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918276133552128 |
|---|---|
| author | Yarkoni, Sol Sharif, Mahmood Livni, Roi |
| author_facet | Yarkoni, Sol Sharif, Mahmood Livni, Roi |
| contents | Recent advances in generative models, such as diffusion models, have raised concerns related to privacy, copyright infringement, and data stewardship. To better understand and control these risks, prior work has introduced techniques and attacks that reconstruct images, or parts of images, from training data. While these results demonstrate that training data can be recovered, existing methods often rely on high computational resources, partial access to the training set, or carefully engineered prompts.
In this work, we present a new attack that requires low resources, assumes little to no access to the training data, and identifies seemingly benign prompts that can lead to potentially risky image reconstruction. We further show that such reconstructions may occur unintentionally, even for users without specialized knowledge. For example, we observe that for one existing model, the prompt ``blue Unisex T-Shirt'' generates the face of a real individual. Moreover, by combining the identified vulnerabilities with real-world prompt data, we discover prompts that reproduce memorized visual elements.
Our approach builds on insights from prior work and leverages domain knowledge to expose a fundamental vulnerability arising from the use of scraped e-commerce data, where templated layouts and images are closely tied to pattern-like textual prompts.
The code for our attack is publicly available at https://github.com/TheSolY/lr-tmi. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_07947 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Low Resource Reconstruction Attacks Through Benign Prompts Yarkoni, Sol Sharif, Mahmood Livni, Roi Machine Learning Artificial Intelligence Recent advances in generative models, such as diffusion models, have raised concerns related to privacy, copyright infringement, and data stewardship. To better understand and control these risks, prior work has introduced techniques and attacks that reconstruct images, or parts of images, from training data. While these results demonstrate that training data can be recovered, existing methods often rely on high computational resources, partial access to the training set, or carefully engineered prompts. In this work, we present a new attack that requires low resources, assumes little to no access to the training data, and identifies seemingly benign prompts that can lead to potentially risky image reconstruction. We further show that such reconstructions may occur unintentionally, even for users without specialized knowledge. For example, we observe that for one existing model, the prompt ``blue Unisex T-Shirt'' generates the face of a real individual. Moreover, by combining the identified vulnerabilities with real-world prompt data, we discover prompts that reproduce memorized visual elements. Our approach builds on insights from prior work and leverages domain knowledge to expose a fundamental vulnerability arising from the use of scraped e-commerce data, where templated layouts and images are closely tied to pattern-like textual prompts. The code for our attack is publicly available at https://github.com/TheSolY/lr-tmi. |
| title | Low Resource Reconstruction Attacks Through Benign Prompts |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2507.07947 |