From Unlearning to UNBRANDING: A Benchmark for Trademark-Safe Text-to-Image Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Malarz, Dawid, Manjak, Filip, Zięba, Maciej, Spurek, Przemysław, Kasymov, Artur
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908919906959360
author Malarz, Dawid
Manjak, Filip
Zięba, Maciej
Spurek, Przemysław
Kasymov, Artur
author_facet Malarz, Dawid
Manjak, Filip
Zięba, Maciej
Spurek, Przemysław
Kasymov, Artur
contents The rapid progress of text-to-image diffusion models raises significant concerns regarding the unauthorized reproduction of trademarked content. While prior work targets general concepts (e.g., styles, celebrities), it fails to address specific brand identifiers. Brand recognition is multi-dimensional, extending beyond explicit logos to encompass distinctive structural features (e.g., a car's front grille). To tackle this, we introduce unbranding, a novel task for the fine-grained removal of both trademarks and subtle structural brand features, while preserving semantic coherence. We construct a benchmark dataset and introduce a novel evaluation framework combining Vision Language Models (VLMs) with segmentation-based classifiers trained on human annotations of logos and trade dress features, addressing the limitations of existing brand detectors that fail to capture abstract trade dress. Furthermore, we observe that newer, higher-fidelity systems (SDXL, FLUX) synthesize brand identifiers more readily than older models, highlighting the urgency of this challenge. Our results confirm that unbranding is a distinct problem requiring specialized techniques. Project Page: https://gmum.github.io/UNBRANDING/.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13953
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Unlearning to UNBRANDING: A Benchmark for Trademark-Safe Text-to-Image Generation
Malarz, Dawid
Manjak, Filip
Zięba, Maciej
Spurek, Przemysław
Kasymov, Artur
Computer Vision and Pattern Recognition
The rapid progress of text-to-image diffusion models raises significant concerns regarding the unauthorized reproduction of trademarked content. While prior work targets general concepts (e.g., styles, celebrities), it fails to address specific brand identifiers. Brand recognition is multi-dimensional, extending beyond explicit logos to encompass distinctive structural features (e.g., a car's front grille). To tackle this, we introduce unbranding, a novel task for the fine-grained removal of both trademarks and subtle structural brand features, while preserving semantic coherence. We construct a benchmark dataset and introduce a novel evaluation framework combining Vision Language Models (VLMs) with segmentation-based classifiers trained on human annotations of logos and trade dress features, addressing the limitations of existing brand detectors that fail to capture abstract trade dress. Furthermore, we observe that newer, higher-fidelity systems (SDXL, FLUX) synthesize brand identifiers more readily than older models, highlighting the urgency of this challenge. Our results confirm that unbranding is a distinct problem requiring specialized techniques. Project Page: https://gmum.github.io/UNBRANDING/.
title From Unlearning to UNBRANDING: A Benchmark for Trademark-Safe Text-to-Image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.13953