# Exploring Local Memorization in Diffusion Models via Bright Ending Attention

arXiv: https://arxiv.org/abs/2410.21665 · HTML: https://mlsanity.com/paper/2410.21665/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 29 Oct 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 19 of 19)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [Large Language Models Struggle to Learn Long-Tail Knowledge](https://mlsanity.com/paper/2211.08411/) (2022)
- [Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models](https://mlsanity.com/paper/2212.03860/) (2022)
- [Deduplicating Training Data Mitigates Privacy Risks in Language Models](https://mlsanity.com/paper/2202.06539/) (2022)
- [Understanding and Mitigating Copying in Diffusion Models](https://mlsanity.com/paper/2305.20086/) (2023)
- [A Self-Supervised Descriptor for Image Copy Detection](https://mlsanity.com/paper/2202.10261/) (2022)
- [Ambient Diffusion: Learning Clean Distributions from Corrupted Data](https://mlsanity.com/paper/2305.19256/) (2023)
- [Detecting, Explaining, and Mitigating Memorization in Diffusion Models](https://mlsanity.com/paper/2407.21720/) (2024)
- [A Reproducible Extraction of Training Images from Diffusion Models](https://mlsanity.com/paper/2305.08694/) (2023)
- [Unveiling and Mitigating Memorization in Text-to-image Diffusion Models through Cross Attention](https://mlsanity.com/paper/2403.11052/) (2024)
- [Towards Memorization-Free Diffusion Models](https://mlsanity.com/paper/2404.00922/) (2024)
- [This Face Does Not Exist ... But It Might Be Yours! Identity Leakage in Generative Models](https://mlsanity.com/paper/2101.05084/) (2020)

## Built on by (most cited first, 19 of 19)

- [Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes](https://mlsanity.com/paper/2412.04140/) (2024)
- [Enhancing Physical Plausibility in Video Generation by Reasoning the Implausibility](https://mlsanity.com/paper/2509.24702/) (2025)
- [Generalization of Diffusion Models Arises with a Balanced Representation Space](https://mlsanity.com/paper/2512.20963/) (2025)
- [Adjusting Initial Noise to Mitigate Memorization in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2510.08625/) (2025)
- [Enhancing Privacy-Utility Trade-offs to Mitigate Memorization in Diffusion Models](https://mlsanity.com/paper/2504.18032/) (2025)
- [Demystifying Foreground-Background Memorization in Diffusion Models](https://mlsanity.com/paper/2508.12148/) (2025)
- [Detecting and Mitigating Memorization in Diffusion Models through Anisotropy of the Log-Probability](https://mlsanity.com/paper/2601.20642/) (2026)
- [Reconstructing Template-Memorized Images from Natural Prompts](https://mlsanity.com/paper/2507.07947/) (2025)
- [Dominant vs. Dominated: Concept-Level Generative Collapse in Diffusion Models](https://mlsanity.com/paper/2512.20666/) (2025)
- [FPAN: Mitigating Replication in Diffusion Models through the Fine-Grained Probabilistic Addition of Noise to Token Embeddings](https://mlsanity.com/paper/2505.21848/) (2025)
- [The Persistence of Cultural Memory: Investigating Multimodal Iconicity in Diffusion Models](https://mlsanity.com/paper/2511.11435/) (2025)
- [CAPTAIN: Semantic Feature Injection for Memorization Mitigation in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2512.10655/) (2025)
- [Beyond Memorization: Selective Learning for Copyright-Safe Diffusion Model Training](https://mlsanity.com/paper/2512.11194/) (2025)
- [Memorization In Stable Diffusion Is Unexpectedly Driven by CLIP Embeddings](https://mlsanity.com/paper/2605.02908/) (2026)
- [Filtering Memorization from Parameter-Space in Diffusion Models](https://mlsanity.com/paper/2605.10439/) (2026)
- [Broken Memories: Detecting and Mitigating Memorization in Diffusion Models with Degraded Generations](https://mlsanity.com/paper/2605.22050/) (2026)
- [Localizing Memorized Regions in Diffusion Models via Coordinate-Wise Curvature Differences](https://mlsanity.com/paper/2605.26756/) (2026)
- [From Modes to Memories: Characterizing the Scale-Space Dynamics of Diffusion Models](https://mlsanity.com/paper/2609.39648/) (2026)
- [Rethinking Memorization Mitigation in Diffusion Models: Reinforcing Text Conditioning](https://mlsanity.com/paper/2610.01723/) (2026)

## Abstract (arXiv)

Text-to-image diffusion models have achieved unprecedented proficiency in generating realistic images. However, their inherent tendency to memorize and replicate training data during inference raises significant concerns, including potential copyright infringement. In response, various methods have been proposed to evaluate, detect, and mitigate memorization. Our analysis reveals that existing approaches significantly underperform in handling local memorization, where only specific image regions are memorized, compared to global memorization, where the entire image is replicated. Also, they cannot locate the local memorization regions, making it hard to investigate locally. To address these, we identify a novel "bright ending" (BE) anomaly in diffusion models prone to memorizing training images. BE refers to a distinct cross-attention pattern observed in text-to-image diffusion models, where memorized image patches exhibit significantly greater attention to the final text token during the last inference step than non-memorized patches. This pattern highlights regions where the generated image replicates training data and enables efficient localization of memorized regions. Equipped with this, we propose a simple yet effective method to integrate BE into existing frameworks, significantly improving their performance by narrowing the performance gap caused by local memorization. Our results not only validate the successful execution of the new localization task but also establish new state-of-the-art performance across all existing tasks, underscoring the significance of the BE phenomenon.