Text-to-Image Models Leave Identifiable Signatures: Implications for Leaderboard Security

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Naseh, Ali, Suri, Anshuman, Peng, Yuefeng, Chaudhari, Harsh, Oprea, Alina, Houmansadr, Amir
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909830931808256
author Naseh, Ali
Suri, Anshuman
Peng, Yuefeng
Chaudhari, Harsh
Oprea, Alina
Houmansadr, Amir
author_facet Naseh, Ali
Suri, Anshuman
Peng, Yuefeng
Chaudhari, Harsh
Oprea, Alina
Houmansadr, Amir
contents Generative AI leaderboards are central to evaluating model capabilities, but remain vulnerable to manipulation. Among key adversarial objectives is rank manipulation, where an attacker must first deanonymize the models behind displayed outputs -- a threat previously demonstrated and explored for large language models (LLMs). We show that this problem can be even more severe for text-to-image leaderboards, where deanonymization is markedly easier. Using over 150,000 generated images from 280 prompts and 19 diverse models spanning multiple organizations, architectures, and sizes, we demonstrate that simple real-time classification in CLIP embedding space identifies the generating model with high accuracy, even without prompt control or historical data. We further introduce a prompt-level separability metric and identify prompts that enable near-perfect deanonymization. Our results indicate that rank manipulation in text-to-image leaderboards is easier than previously recognized, underscoring the need for stronger defenses.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06525
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Text-to-Image Models Leave Identifiable Signatures: Implications for Leaderboard Security
Naseh, Ali
Suri, Anshuman
Peng, Yuefeng
Chaudhari, Harsh
Oprea, Alina
Houmansadr, Amir
Machine Learning
Cryptography and Security
Generative AI leaderboards are central to evaluating model capabilities, but remain vulnerable to manipulation. Among key adversarial objectives is rank manipulation, where an attacker must first deanonymize the models behind displayed outputs -- a threat previously demonstrated and explored for large language models (LLMs). We show that this problem can be even more severe for text-to-image leaderboards, where deanonymization is markedly easier. Using over 150,000 generated images from 280 prompts and 19 diverse models spanning multiple organizations, architectures, and sizes, we demonstrate that simple real-time classification in CLIP embedding space identifies the generating model with high accuracy, even without prompt control or historical data. We further introduce a prompt-level separability metric and identify prompts that enable near-perfect deanonymization. Our results indicate that rank manipulation in text-to-image leaderboards is easier than previously recognized, underscoring the need for stronger defenses.
title Text-to-Image Models Leave Identifiable Signatures: Implications for Leaderboard Security
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2510.06525