# DIAGNOSIS: Detecting Unauthorized Data Usages in Text-to-image Diffusion Models Source: [DIAGNOSIS: Detecting Unauthorized Data Usages in Text-to-image Diffusion Models]() Collection index: [llms.txt]() Data updated: 2025-10-27T06:32:05.519Z Authors: Zhenting Wang, Chen Chen, Lingjuan Lyu, Dimitris N. Metaxas, Shiqing Ma Published: 2023-07-06T16:27:39.000Z ## Abstract Recent text-to-image diffusion models have shown surprising performance in generating high-quality images. However, concerns have arisen regarding the unauthorized data usage during the training or fine-tuning process. One example is when a model trainer collects a set of images created by a particular artist and attempts to train a model capable of generating similar images without obtaining permission and giving credit to the artist. To address this issue, we propose a method for detecting such unauthorized data usage by planting the injected memorization into the text-to-image diffusion models trained on the protected dataset. Specifically, we modify the protected images by adding unique contents on these images using stealthy image warping functions that are nearly imperceptible to humans but can be captured and memorized by diffusion models. By analyzing whether the model has memorized the injected content (i.e., whether the generated images are processed by the injected post-processing function), we can detect models that had illegally utilized the unauthorized data. Experiments on Stable Diffusion and VQ Diffusion with different model training or fine-tuning methods (i.e, LoRA, DreamBooth, and standard training) demonstrate the effectiveness of our proposed method in detecting unauthorized data usages. Code: https://github.com/ZhentingWang/DIAGNOSIS. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Watermark Detection on Pokemon](): Watermark Detection; Pokemon; Pu (R=0): 0. - [Watermark Detection on FFHQ](): Watermark Detection; FFHQ; Pu (R=0): 3.8. - [Semantic Watermark Detection on CelebA-HQ protected DreamBooth-generated](): Semantic Watermark Detection; CelebA-HQ protected DreamBooth-generated; Detection Percentage: 17.24. - [Watermark Detection on CUB-200](): Watermark Detection; CUB-200; Pu (R=0): 6. - [Image Quality Assessment on ImageNet (PSNR/SSIM)](): Image Quality Assessment; ImageNet; PSNR: 32.21. - [Image Quality Assessment on LSUN](): Image Quality Assessment; LSUN; PSNR: 32.84. - [Image Quality Assessment on Pokemon](): Image Quality Assessment; Pokemon; PSNR: 27.04. - [Image Quality Assessment on FFHQ](): Image Quality Assessment; FFHQ; PSNR: 32.45. - [Infringement Model Detection on Pokemon](): Infringement Model Detection; Pokemon; TP: 10. ## Original sources and implementation [Original paper]() [PDF]()