mlsanityData to 7 Oct 2026

ProMark: Proactive Diffusion Watermarking for Causal Attribution

arXiv 2403.09914 ↗First posted 14 Mar 2024cs.CVcited by 16

On the map: LLM safety and security → Watermarking generated content and models

Abstract (arXiv)

Generative AI (GenAI) is transforming creative workflows through the capability to synthesize and manipulate images via high-level prompts. Yet creatives are not well supported to receive recognition or reward for the use of their content in GenAI training. To this end, we propose ProMark, a causal attribution technique to attribute a synthetically generated image to its training data concepts like objects, motifs, templates, artists, or styles. The concept information is proactively embedded into the input training images using imperceptible watermarks, and the diffusion models (unconditional or conditional) are trained to retain the corresponding watermarks in generated images. We show that we can embed as many as 216 unique watermarks into the training data, and each training image can contain more than one watermark. ProMark can maintain image quality whilst outperforming correlation-based attribution. Finally, several qualitative examples are presented, providing the confidence that the presence of the watermark conveys a causative relationship between training data and synthetic images.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security. Builds on 31 of its 39 references that are in the map. First posted on arXiv 14 Mar 2024.

Builds on

31 of 39 references.

Closest ancestors · 7

Same direction, newest first

  1. 2023Invisible Watermarking for Audio Generation Diffusion Models22 Sep
  2. 2023DiffusionShield: A Watermark for Copyright Protection against Generative Diffusion Models25 May
  3. 2023Watermarking Diffusion Model21 May
  4. 2023RoSteALS: Robust Steganography using Autoencoder Latent Space6 Apr
  5. 2023The Stable Signature: Rooting Watermarks in Latent Diffusion Models27 Mar
  6. 2023A Recipe for Watermarking Diffusion Models17 Mar
  7. 2023A Watermark for Large Language Models24 Jan

Other references · 24

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  5. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  6. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 847
  7. 2022On Distillation of Guided Diffusion Modelscited by 721
  8. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 492
  9. 2023Ablating Concepts in Text-to-Image Diffusion Modelscited by 312
  10. 2022A Self-Supervised Descriptor for Image Copy Detectioncited by 175
  11. 2020Radioactive data: tracing through trainingcited by 67
  12. 2020Disrupting Deepfakes: Adversarial Attacks Against Conditional Image Translation Networks and Facial Manipulation Systemscited by 65
12 more
  1. 2021Reverse Engineering of Generative Models: Inferring Model Hyperparameters from Generated Imagescited by 58
  2. 2020FakeTagger: Robust Safeguards against DeepFake Dissemination via Provenance Trackingcited by 46
  3. 2023Evaluating Data Attribution for Text-to-Image Modelscited by 45
  4. 2017BAM! The Behance Artistic Media Dataset for Recognition Beyond Photographycited by 43
  5. 2022RepMix: Representation Mixing for Robust Attribution of Synthesized Imagescited by 39
  6. 2022Proactive Image Manipulation Detectioncited by 35
  7. 2021ALADIN: All Layer Adaptive Instance Normalization for Fine-grained Style Similaritycited by 34
  8. 2023MaLP: Manipulation Localization Using a Proactive Schemecited by 26
  9. 2021OSCAR-Net: Object-centric Scene Graph Attention for Image Attributioncited by 15
  10. 2023EKILA: Synthetic Media Provenance and Attribution for Generative Artcited by 11
  11. 2020Camera Trace Erasingcited by 7
  12. 2023PrObeD: Proactive Object Detection Wrappercited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 8; 2026: 5.

16 papers.

Sort:

By area:

  1. 2024OmniGuard: Hybrid Manipulation Localization via Augmented Versatile Deep Image Watermarkingcited by 35
  2. 2025Semantic Watermarking Reinvented: Enhancing Robustness and Generation Quality with Fourier Integritycited by 14
  3. 2025On the Coexistence and Ensembling of Watermarkscited by 3
  4. 2025Content ARCs: Decentralized Content Rights in the Age of Generative AIcited by 3
  5. 2025Attribution-by-design: Ensuring Inference-Time Provenance in Generative Music Systemscited by 3
  6. 2024Perturb, Attend, Detect and Localize (PADL): Robust Proactive Image Defensecited by 1
  7. 2024An Efficient Watermarking Method for Latent Diffusion Models via Low-Rank Adaptation and Dynamic Loss Weightingcited by 1
  8. 2025KGMark: A Diffusion Watermark for Knowledge Graphscited by 1
  9. 2026Learning Probabilistic Embeddings for Unsupervised Action Segmentationcited by 1
  10. 2025MultiNeRF: Multiple Watermark Embedding for Neural Radiance Fieldscited by 0
  11. 2025Protecting Your Voice: Temporal-aware Robust Watermarkingcited by 0
  12. 2025TriniMark: A Robust Generative Speech Watermarking Method for Trinity-Level Traceabilitycited by 0
4 more
  1. 2026TokenTrace: Multi-Concept Attribution through Watermarked Token Recoverycited by 0
  2. 2026Splats in Splats++: Robust and Generalizable 3D Gaussian Splatting Steganographycited by 0
  3. 2026Membership is Ownership: A Robust Ownership Verification Framework for Diffusion Modelscited by 0
  4. 2026FeatMark: Feature-level Watermark Protection against Mimicry Attacks with Diffusion Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Image Watermarking of Generative Diffusion Models
  2. 2023Hey That's Mine Imperceptible Watermarks are Preserved in Diffusion Generated Outputs
  3. 2025MarkDiffusion: An Open-Source Toolkit for Generative Watermarking of Latent Diffusion Models
  4. 2024WMAdapter: Adding WaterMark Control to Latent Diffusion Models
  5. 2025HMARK: Radioactive Multi-Bit Semantic-Latent Watermarking for Diffusion Models
  6. 2025Radioactive Watermarks in Diffusion and Autoregressive Image Generative Models
  7. 2025TraceMark-LDM: Authenticatable Watermarking for Latent Diffusion Models via Binary-Guided Rearrangement
  8. 2024DiffuseTrace: A Transparent and Flexible Watermarking Scheme for Latent Diffusion Model