Research map / Image generation / Autoregressive and latent generation
PlugMark: A Plug-in Zero-Watermarking Framework for Diffusion Models
Pengzhen Chen, Yanwei Liu, Xiaoyan Gu, Enci Liu, Zhuoyi Shang, Xiangyang Ji, Wu Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add image+text input to DKe training instead of image only (ablation) | PlugMark · fine-tuned 1000 steps | matching accuracy (%): 40.67 -> 66.09 | Improved |
| generate multimodal (image, text) boundary pairs instead of image-only BRoK (ablation) | PlugMark · fine-tuned 1000 steps | matching accuracy (%): 66.09 -> 85.18 | Improved |
| add multi-timestep features to DKe (ablation) | PlugMark · fine-tuned 1000 steps | matching accuracy (%): 85.18 -> 94.82 | Improved |
| add uncertainty-based (variance) timestep selection and weighting (ablation) | PlugMark · fine-tuned 1000 steps | matching accuracy (%): 94.82 -> 98.83 | Improved |
| add multi-timestep features to DKe (ablation) | PlugMark · pruned rho=0.1 | matching accuracy (%): 90.00 -> 94.90 | Improved |
| add uncertainty-based timestep selection (ablation) | PlugMark · pruned rho=0.1 | matching accuracy (%): 94.90 -> 98.00 | Improved |
| use PlugMark boundary generation instead of IPguard (ablation) | BRoK · fine-tune 2000 steps | TPR: 0.005 -> 0.991 | Improved |
| use zero-watermarking PlugMark instead of Stable Signature | Stable Diffusion v2 · FID | FID: 24.77 -> 24.25 | Improved |
| use PlugMark instead of AquaLoRA | Stable Diffusion v2 · fine-tune 1000 steps | TPR at FPR=1e-6: 0.003 -> 0.995 | Improved |
| use CIFAR-100 (20 classes) instead of CIFAR-10 for DKe | PlugMark · fine-tuned 20k steps | matching accuracy (%): 72.55 -> 69.18 | Hurt |
| increase number of selected timesteps k (ablation) | PlugMark · fine-tuned 1000 steps | matching accuracy | Improved |
| select timesteps by lowest variance instead of randomly (ablation) | PlugMark · fine-tuned 1000 steps | matching accuracy | Improved |
Closest papers by meaning
- SleeperMark: Towards Robust Watermark against Fine-Tuning Text-to-image Diffusion Models (CVPR 2025 · similarity 0.91)
- Harnessing Frequency Spectrum Insights for Image Copyright Protection Against Diffusion Models (CVPR 2025 · similarity 0.91)
- AquaLoRA: Toward White-box Protection for Customized Stable Diffusion Models via Watermark LoRA (ICML 2024 · similarity 0.89)
- Shallow Diffuse: Robust and Invisible Watermarking through Low-Dim Subspaces in Diffusion Models (NeurIPS 2025 · similarity 0.89)
- WMAdapter: Adding WaterMark Control to Latent Diffusion Models (ICML 2025 · similarity 0.89)
- Black-Box Forgery Attacks on Semantic Watermarks for Diffusion Models (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
