MemBench: Memorized Image Trigger Prompt Dataset for Diffusion Models
Keywords: diffusion models, MCMC sampling, image memorization, safety ai
TL;DR: We introduce a benchmark containing prompts that induce memorized images in text-to-image diffusion models. Using this benchmark, we evaluate various memorization mitigation methods.
Abstract: Diffusion models have achieved remarkable success in Text-to-Image generation tasks, leading to the development of many commercial models. However, recent studies have reported that diffusion models often repeatedly generate memorized images in train data when triggered by specific prompts, potentially raising social issues ranging from copyright to privacy concerns. To sidestep the memorization, there have been recent studies for developing memorization mitigation methods for diffusion models. Nevertheless, the lack of benchmarks hinders the assessment of the true effectiveness of these methods. In this work, we present MemBench, the first benchmark for evaluating image memorization mitigation methods. Our benchmark includes a large number of memorized image trigger prompts in various Text-to-Image diffusion models. Furthermore, in contrast to the prior work evaluating mitigation performance only on trigger prompts, we present metrics evaluating on both trigger prompts and general prompts, so that we can see whether mitigation methods address the memorization issue while maintaining performance for general prompts. Through our MemBench evaluation, we revealed that existing memorization mitigation methods notably degrade overall performance of diffusion models and need to be further developed.
Supplementary Material: zip
Primary Area: generative models
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 6571
Loading