mlsanityData to 14 Oct 2026

T2IW: Joint Text to Image & Watermark Generation

arXiv 2309.03815 ↗First posted 7 Sep 2023cs.CV cs.MM eess.IVcited by 5

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

Abstract (arXiv)

Recent developments in text-conditioned image generative models have revolutionized the production of realistic results. Unfortunately, this has also led to an increase in privacy violations and the spread of false information, which requires the need for traceability, privacy protection, and other security measures. However, existing text-to-image paradigms lack the technical capabilities to link traceable messages with image generation. In this study, we introduce a novel task for the joint generation of text to image and watermark (T2IW). This T2IW scheme ensures minimal damage to image quality when generating a compound image by forcing the semantic feature and the watermark signal to be compatible in pixels. Additionally, by utilizing principles from Shannon information theory and non-cooperative game theory, we are able to separate the revealed image and the revealed watermark from the compound image. Furthermore, we strengthen the watermark robustness of our approach by subjecting the compound image to various post-processing attacks, with minimal pixel distortion observed in the revealed watermark. Extensive experiments have demonstrated remarkable achievements in image quality, watermark invisibility, and watermark robustness, supported by our proposed set of evaluation metrics.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security. Builds on 32 of its 84 references that are in the map. First posted on arXiv 7 Sep 2023.

Builds on

32 of 84 references.

References · 32

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  6. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  7. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  8. 2016Improved Techniques for Training GANscited by 2,970
  9. 2016Generative Adversarial Text to Image Synthesiscited by 735
  10. 2017AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networkscited by 664
  11. 2016StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 577
  12. 2018HiDDeN: Hiding Data With Deep Networkscited by 397
20 more
  1. 2019DM-GAN: Dynamic Memory Generative Adversarial Networks for Text-to-Image Synthesiscited by 308
  2. 2019StegaStamp: Invisible Hyperlinks in Physical Photographscited by 265
  3. 2017StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 229
  4. 2020Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Datacited by 201
  5. 2020DF-GAN: A Simple and Effective Baseline for Text-to-Image Synthesiscited by 162
  6. 2019Object-driven Text-to-Image Synthesis via Adversarial Trainingcited by 117
  7. 2016Learning What and Where to Drawcited by 105
  8. 2018Image Generation from Layoutcited by 105
  9. 2019Semantic Object Accuracy for Generative Text-to-Image Synthesiscited by 89
  10. 2017Semantic Image Synthesis via Adversarial Learningcited by 87
  11. 2023GALIP: Generative Adversarial CLIPs for Text-to-Image Synthesiscited by 85
  12. 2019SteganoGAN: High Capacity Image Steganography with GANscited by 72
  13. 2019Semantics Disentangling for Text-to-Image Generationcited by 61
  14. 2021Protecting Intellectual Property of Generative Adversarial Networks from Ambiguity Attackcited by 46
  15. 2019Generating Multiple Objects at Spatially Distinct Locationscited by 41
  16. 2018Adversarial Learning of Semantic Relevance in Text to Image Synthesiscited by 14
  17. 2020DTGAN: Dual Attention Generative Adversarial Networks for Text-to-Image Generationcited by 14
  18. 2022Recurrent Affine Transformation for Text-to-image Synthesiscited by 11
  19. 2022CLIP2GAN: Towards Bridging Text with the Latent Space of GANscited by 8
  20. 2023A Comprehensive Review and Systematic Analysis of Artificial Intelligence Regulation Policiescited by 4

Built on by

Cited by year

Citing papers per year: 2024: 3; 2025: 2.

5 papers.

  1. 2024Watermarking Techniques for Large Language Models: A Surveycited by 23
  2. 2024SoK: On the Role and Future of AIGC Watermarking in the Era of Gen-AIcited by 14
  3. 2025Survey on AI-Generated Media Detection: From Non-MLLM to MLLMcited by 12
  4. 2024MarkPlugger: Generalizable Watermark Framework for Latent Diffusion Models without Retrainingcited by 5
  5. 2025PCDiff: Proactive Control for Ownership Protection in Diffusion Models with Watermark Compatibilitycited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2024Exploiting Watermark-Based Defense Mechanisms in Text-to-Image Diffusion Models for Unauthorized Data Usage
  2. 2025BitMark: Watermarking Bitwise Autoregressive Image Generative Models
  3. 2024Robustness of Watermarking on Text-to-Image Diffusion Models
  4. 2026Persistent Watermarking of Text-to-Image Models
  5. 2025CEFW: A Comprehensive Evaluation Framework for Watermark in Large Language Models
  6. 2024Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach
  7. 2026On Google's SynthID-Text LLM Watermarking System: Theoretical Analysis and Empirical Validation
  8. 2025Secure and Robust Watermarking for AI-generated Images: A Comprehensive Survey