Identifying Models Behind Text-to-Image Leaderboards
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918289305763840 |
|---|---|
| author | Naseh, Ali Peng, Yuefeng Suri, Anshuman Chaudhari, Harsh Oprea, Alina Houmansadr, Amir |
| author_facet | Naseh, Ali Peng, Yuefeng Suri, Anshuman Chaudhari, Harsh Oprea, Alina Houmansadr, Amir |
| contents | Text-to-image (T2I) models are increasingly popular, producing a large share of AI-generated images online. To compare model quality, voting-based leaderboards have become the standard, relying on anonymized model outputs for fairness. In this work, we show that such anonymity can be easily broken. We find that generations from each T2I model form distinctive clusters in the image embedding space, enabling accurate deanonymization without prompt control or training data. Using 22 models and 280 prompts (150K images), our centroid-based method achieves high accuracy and reveals systematic model-specific signatures. We further introduce a prompt-level distinguishability metric and conduct large-scale analyses showing how certain prompts can lead to near-perfect distinguishability. Our findings expose fundamental security flaws in T2I leaderboards and motivate stronger anonymization defenses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_09647 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Identifying Models Behind Text-to-Image Leaderboards Naseh, Ali Peng, Yuefeng Suri, Anshuman Chaudhari, Harsh Oprea, Alina Houmansadr, Amir Computer Vision and Pattern Recognition Cryptography and Security Machine Learning Text-to-image (T2I) models are increasingly popular, producing a large share of AI-generated images online. To compare model quality, voting-based leaderboards have become the standard, relying on anonymized model outputs for fairness. In this work, we show that such anonymity can be easily broken. We find that generations from each T2I model form distinctive clusters in the image embedding space, enabling accurate deanonymization without prompt control or training data. Using 22 models and 280 prompts (150K images), our centroid-based method achieves high accuracy and reveals systematic model-specific signatures. We further introduce a prompt-level distinguishability metric and conduct large-scale analyses showing how certain prompts can lead to near-perfect distinguishability. Our findings expose fundamental security flaws in T2I leaderboards and motivate stronger anonymization defenses. |
| title | Identifying Models Behind Text-to-Image Leaderboards |
| topic | Computer Vision and Pattern Recognition Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2601.09647 |