mlsanityData to 7 Oct 2026

Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hiding

arXiv 2009.03015 ↗First posted 7 Sep 2020cs.CR cs.CL cs.CYcited by 121

Not on the map. Why?

Cited by 69 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Recent advances in natural language generation have introduced powerful language models with high-quality output text. However, this raises concerns about the potential misuse of such models for malicious purposes. In this paper, we study natural language watermarking as a defense to help better mark and trace the provenance of text. We introduce the Adversarial Watermarking Transformer (AWT) with a jointly trained encoder-decoder and adversarial training that, given an input text and a binary message, generates an output text that is unobtrusively encoded with the given message. We further study different training and inference strategies to achieve minimal changes to the semantics and correctness of the input text. AWT is the first end-to-end model to hide data in text by automatically learning -- without ground truth -- word substitutions along with their locations in order to encode the message. We empirically show that our model is effective in largely preserving text utility and decoding the watermark while hiding its presence against adversaries. Additionally, we demonstrate that our method is robust against a range of attacks.

Key facts. Cited by 121 papers in the map (data through 7 Oct 2026). First posted on arXiv 7 Sep 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 5; 2023: 38; 2024: 35; 2025: 24; 2026: 13.

121 papers.

Sort:

By area:

  1. 2023A Watermark for Large Language Modelscited by 667
  2. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 401
  3. 2023The Stable Signature: Rooting Watermarks in Latent Diffusion Modelscited by 289
  4. 2023Robust Distortion-free Watermarks for Language Modelscited by 279
  5. 2023Fast-DetectGPT: Efficient Zero-Shot Detection of Machine-Generated Text via Conditional Probability Curvaturecited by 278
  6. 2023Undetectable Watermarks for Language Modelscited by 208
  7. 2023DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of Machine-Generated Textcited by 207
  8. 2023On the Reliability of Watermarks for Large Language Modelscited by 170
  9. 2023The Science of Detecting LLM-Generated Textscited by 126
  10. 2023Who Wrote this Code? Watermarking for Code Generationcited by 116
  11. 2023A Survey of Text Watermarking in the Era of Large Language Modelscited by 102
  12. 2023A Semantic Invariant Robust Watermark for Large Language Modelscited by 97
38 more
  1. 2022Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protectioncited by 80
  2. 2024Watermark Stealing in Large Language Modelscited by 79
  3. 2023Robust Multi-bit Natural Language Watermarking through Invariant Featurescited by 76
  4. 2023Evading Watermark based Detection of AI-Generated Contentcited by 74
  5. 2023On the Learnability of Watermarks for Language Modelscited by 72
  6. 2023A Survey on Detection of LLMs-Generated Contentcited by 70
  7. 2023Publicly-Detectable Watermarking for Language Modelscited by 69
  8. 2023Watermarking Text Generated by Black-Box Language Modelscited by 68
  9. 2023REMARK-LLM: A Robust and Efficient Watermarking Framework for Generative Large Language Modelscited by 64
  10. 2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systemscited by 63
  11. 2024Adaptive Text Watermark for Large Language Modelscited by 60
  12. 2022"Real Attackers Don't Compute Gradients": Bridging the Gap Between Adversarial ML Research and Practicecited by 55
  13. 2023An Unforgeable Publicly Verifiable Watermark for Large Language Modelscited by 54
  14. 2021Changing the World by Changing the Datacited by 51
  15. 2023Advancing Beyond Identification: Multi-bit Watermark for Large Language Modelscited by 51
  16. 2023A Resilient and Accessible Distribution-Preserving Watermark for Large Language Modelscited by 47
  17. 2023LLMDet: A Third Party Large Language Models Generated Text Detection Toolcited by 44
  18. 2023Large Language Models can be Guided to Evade AI-Generated Text Detectioncited by 43
  19. 2023Towards Codable Watermarking for Injecting Multi-bits Information to LLMscited by 43
  20. 2021Tracing Text Provenance via Context-Aware Lexical Substitutioncited by 40
  21. 2022Attribution and Obfuscation of Neural Text Authorship: A Data Mining Perspectivecited by 40
  22. 2024AI as Humanity's Salieri: Quantifying Linguistic Creativity of Language Models via Systematic Attribution of Machine Text against Web Textcited by 40
  23. 2023Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitationcited by 37
  24. 2023Smaller Language Models are Better Black-box Machine-Generated Text Detectorscited by 35
  25. 2023A Survey on ChatGPT: AI-Generated Contents, Challenges, and Solutionscited by 34
  26. 2024TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identificationcited by 33
  27. 2023HuRef: HUman-REadable Fingerprint for Large Language Modelscited by 32
  28. 2024Token-Specific Watermarking with Enhanced Detectability and Semantic Coherence for Large Language Modelscited by 30
  29. 2025Towards Trustworthy Retrieval Augmented Generation for Large Language Models: A Surveycited by 29
  30. 2021Frustratingly Easy Edit-based Linguistic Steganography with a Masked Language Modelcited by 28
  31. 2024A Transfer Attack to Image Watermarkscited by 26
  32. 2023Towards Optimal Statistical Watermarkingcited by 25
  33. 2024PostMark: A Robust Blackbox Watermark for Large Language Modelscited by 24
  34. 2021Data Hiding with Deep Learning: A Survey Unifying Digital Watermarking and Steganographycited by 23
  35. 2023Embarrassingly Simple Text Watermarkscited by 23
  36. 2023Optimizing watermarks for large language modelscited by 22
  37. 2024Watermarking Techniques for Large Language Models: A Surveycited by 20
  38. 2025BiMark: Unbiased Multilayer Watermarking for Large Language Modelscited by 18

50 most cited of 121 All 121 (JSON)