Research map / Image generation / Text-to-image diffusion
Scalable Dual Fingerprinting for Hierarchical Attribution of Text-to-Image Models
Jianwei Fei, Yunshu Dai, Peipeng Yu, Zhe Kong, Jiantao Zhou, Zhihua Xia
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 attack simulation (noise) layer during fine-tuning (ablation) | FI-LoRA · JPEG compression or Gaussian noise (Std=0.2) | Bit Acc: 50% -> nearly 100% | Improved |
| use FI-LoRA instead of WOUAF for image fidelity | SD2-base fingerprinting · COCO | PSNR_b: 31.345 -> 32.359 | Improved |
| use FI-LoRA instead of baselines for generation quality | SD2-base fingerprinting · COCO | FID_r: 23.726 -> 23.272 | Improved |
| embed two fingerprints (64 bits total) instead of one | FI-LoRA · COCO | Bit Acc (%): 99.67 (Per. Norm.) -> 99.96 / 99.91 | No clear effect |
| apply 50% pruning or noise std 0.05 model attack | FI-LoRA · model modification attacks | Bit Acc: about 5% below Per. Norm. | Hurt |
| adversarial fine-tuning of the model for 10k steps | FI-LoRA · model fine-tuning attack | Bit Acc: remains above 80% vs below 70% for baselines | Mixed |
| use a single 64-bit fingerprint instead of two 32-bit fingerprints | FI-LoRA · capacity limit | convergence | Hurt |
| 2x brightness attack on generated images | FI-LoRA · image-level attacks | Bit Acc: below 80% with PSNR 10 or lower | Hurt |
Closest papers by meaning
- WOUAF: Weight Modulation for User Attribution and Fingerprinting in Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.88)
- Scalable Fingerprinting of Large Language Models (NeurIPS 2025 · similarity 0.85)
- SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models (CVPR 2026 · similarity 0.84)
- ManiFPT: Defining and Analyzing Fingerprints of Generative Models (CVPR 2024 · similarity 0.84)
- Fingerprinting Denoising Diffusion Probabilistic Models (CVPR 2025 · similarity 0.83)
- OmniMark: Efficient and Scalable Latent Diffusion Model Fingerprinting (AAAI 2025 · similarity 0.83)
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.
