Spaces:
Running
on
T4
Apply for community grant: Academic project
WOUAF: Weight Modulation for User Attribution and Fingerprinting in Text-to-Image Diffusion Models
Overview
The rapid advancement of generative models, facilitating the creation of hyper-realistic images from textual descriptions, has concurrently escalated critical societal concerns such as misinformation. Traditional fake detection mechanisms, although providing some mitigation, fall short in attributing responsibility for the malicious use of synthetic images.
To approach such social misinformation concerns, we present a novel scalable approach to attribute diffusion models. WOUAF enables the integration of up to 32-bit (~4 billion) fingerprints into Text-to-Image Diffusion Models without loss in image quality.
We want to open-source this demo on the Text-to-Image (T2I) application of GenAI. We hope this demo will inspire future work in the safety of open-source diffusion models.
Here is the final working demo screenshot for reference:
Hi @mpatel57 , we have assigned a gpu to this space. Note that GPU Grants are provided temporarily and might be removed after some time if the usage is very low.
To learn more about GPUs in Spaces, please check out https://huggingface.co/docs/hub/spaces-gpus