# LU-500: A Logo Benchmark for Concept Unlearning

arXiv: https://arxiv.org/abs/2607.24101 · HTML: https://mlsanity.com/paper/2607.24101/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Vision-language and text-to-image safety, one of 12 directions in LLM safety and security.
- Builds on 34 of its 52 references that are in the map.
- First posted on arXiv 27 Jul 2026.

## Placed in

- LLM safety and security / [Vision-language and text-to-image safety](https://mlsanity.com/safety/vision-language-and-text-to-image-safety/)

## Builds on (most cited first, 30 of 34)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2303.07345/) (2023)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [Unified Concept Editing in Diffusion Models](https://mlsanity.com/paper/2308.14761/) (2023)
- [Scaling Open-Vocabulary Object Detection](https://mlsanity.com/paper/2306.09683/) (2023)
- [Ablating Concepts in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.13516/) (2023)
- [A Survey of Machine Unlearning](https://mlsanity.com/paper/2209.02299/) (2022)
- [MACE: Mass Concept Erasure in Diffusion Models](https://mlsanity.com/paper/2403.06135/) (2024)
- [Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.17591/) (2023)
- [One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications](https://mlsanity.com/paper/2312.16145/) (2023)
- [Unlearning Concepts in Diffusion Model via Concept Domain Correction and Concept Preserving Gradient](https://mlsanity.com/paper/2405.15304/) (2024)
- [Towards Safe Self-Distillation of Internet-Scale Text-to-Image Diffusion Models](https://mlsanity.com/paper/2307.05977/) (2023)
- [Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts](https://mlsanity.com/paper/2410.12777/) (2024)
- [Copyright Traps for Large Language Models](https://mlsanity.com/paper/2402.09363/) (2024)
- [SHIELD: Evaluation and Defense Strategies for Copyright Compliance in LLM Text Generation](https://mlsanity.com/paper/2406.12975/) (2024)
- [AgencyBench: Benchmarking the Frontiers of Autonomous Agents in 1M-Token Real-World Contexts](https://mlsanity.com/paper/2601.11044/) (2026)
- [LogoDet-3K: A Large-Scale Image Dataset for Logo Detection](https://mlsanity.com/paper/2008.05359/) (2020)
- [Logo-2K+: A Large-Scale Logo Dataset for Scalable Logo Classification](https://mlsanity.com/paper/1911.07924/) (2019)
- [InnovatorBench: Evaluating Agents' Ability to Conduct Innovative LLM Research](https://mlsanity.com/paper/2510.27598/) (2025)
- [LIMI: Less is More for Agency](https://mlsanity.com/paper/2509.17567/) (2025)
- [Editing Massive Concepts in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2403.13807/) (2024)
- [Open Set Logo Detection and Retrieval](https://mlsanity.com/paper/1710.10891/) (2017)
- [Open Logo Detection Challenge](https://mlsanity.com/paper/1807.01964/) (2018)
- [LOGO-Net: Large-scale Deep Logo Detection and Brand Recognition with Deep Region-based Convolutional Networks](https://mlsanity.com/paper/1511.02462/) (2015)
- [FoodLogoDet-1500: A Dataset for Large-Scale Food Logo Detection via Multi-Scale Feature Decoupling Network](https://mlsanity.com/paper/2108.04644/) (2021)
- [DatasetResearch: Benchmarking Agent Systems for Demand-Driven Dataset Discovery](https://mlsanity.com/paper/2508.06960/) (2025)
- [Deep Learning for Logo Detection: A Survey](https://mlsanity.com/paper/2210.04399/) (2022)
- [The Open Brands Dataset: Unified brand detection and recognition at scale](https://mlsanity.com/paper/2012.07350/) (2020)

## Abstract (arXiv)

Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models. Existing evaluations, however, mostly study targets that dominate the whole image, such as styles, broad object categories, or portrait-like identities, leaving company logos comparatively underexamined. Logos create a different failure mode: a small localized mark can carry the entire protected concept, must be visually precise to remain recognizable, and can be triggered implicitly by products, storefronts, packaging, or advertisements even when the word ``logo'' is absent. We introduce LU-500, a logo-unlearning benchmark built from Fortune Global 500 companies to study this localized and semantically entangled setting. LU-500 contains nearly 10,000 curated text-query and logo-image pairs, with an explicit track (LUex-500) and an implicit contextual track (LUim-500). To avoid reducing the task to a binary detector score, we define a multi-grained protocol that evaluates both local logo removal and global image preservation in pixel and latent spaces. Experiments on representative inference-time methods, including NP, SLD, and SEGA, and compatible fine-tuning-based methods such as ESD and Forget-Me-Not, show that the evaluated methods struggle to remove logo evidence without changing non-target content. We further analyze ProLU, a prompt-space multi-agent baseline: it improves local erasure by removing logo-inducing semantics, but also illustrates why prompt filtering is not a substitute for weight-level disentanglement. Correlation analyses over logo area, location, and structural complexity suggest that future logo unlearning may need spatially aware controls, such as SSIM-guided constraints, rather than purely global concept suppression.