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

2025-12-15

AI Summary

This work addresses unauthorized trademark replication in text-to-image generation by introducing “unbranding”—a novel task that fine-grainedly removes both explicit trademarks (e.g., logos) and implicit brand cues (e.g., car grilles, bottle silhouettes) while preserving semantic fidelity. Methodologically, we formally define multi-dimensional brand identification and suppression; construct the first benchmark dataset for trademark-safe image generation; and propose a vision-language model (VLM)-based QA metric to uniformly evaluate explicit marks and abstract trade dress. Our approach integrates diffusion models (SDXL/FLUX), fine-grained feature disentanglement, and controllable generation. Experiments demonstrate that unbranding constitutes a distinct, practically relevant challenge; higher-fidelity generative models exhibit greater trademark leakage; and our method significantly reduces brand identifiability while achieving superior semantic preservation over existing baselines.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

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. Crucially, we note that 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. To facilitate research, we construct a comprehensive benchmark dataset. Recognizing that existing brand detectors are limited to logos and fail to capture abstract trade dress (e.g., the shape of a Coca-Cola bottle), we introduce a novel evaluation metric based on Vision Language Models (VLMs). This VLM-based metric uses a question-answering framework to probe images for both explicit logos and implicit, holistic brand characteristics. Furthermore, we observe that as model fidelity increases, with newer systems (SDXL, FLUX) synthesizing brand identifiers more readily than older models (Stable Diffusion), the urgency of the unbranding challenge is starkly highlighted. Our results, validated by our VLM metric, confirm unbranding is a distinct, practically relevant problem requiring specialized techniques. Project Page: https://gmum.github.io/UNBRANDING/.

Problem

• Addresses unauthorized trademark reproduction in text-to-image models
• Removes both explicit logos and subtle structural brand features
• Introduces a benchmark and VLM-based metric for brand detection

Innovation

• Introduces unbranding for removing trademarks and structural brand features
• Uses Vision Language Models for evaluating explicit and implicit brand characteristics
• Constructs benchmark dataset for fine-grained trademark-safe image generation