mlsanityData to 7 Oct 2026

Unbiased Watermark for Large Language Models

arXiv 2310.10669 ↗First posted 22 Sep 2023cs.CRcited by 100

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

Abstract (arXiv)

The recent advancements in large language models (LLMs) have sparked a growing apprehension regarding the potential misuse. One approach to mitigating this risk is to incorporate watermarking techniques into LLMs, allowing for the tracking and attribution of model outputs. This study examines a crucial aspect of watermarking: how significantly watermarks impact the quality of model-generated outputs. Previous studies have suggested a trade-off between watermark strength and output quality. However, our research demonstrates that it is possible to integrate watermarks without affecting the output probability distribution with appropriate implementation. We refer to this type of watermark as an unbiased watermark. This has significant implications for the use of LLMs, as it becomes impossible for users to discern whether a service provider has incorporated watermarks or not. Furthermore, the presence of watermarks does not compromise the performance of the model in downstream tasks, ensuring that the overall utility of the language model is preserved. Our findings contribute to the ongoing discussion around responsible AI development, suggesting that unbiased watermarks can serve as an effective means of tracking and attributing model outputs without sacrificing output quality.

Key facts. Cited by 100 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. First posted on arXiv 22 Sep 2023.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 30; 2025: 34; 2026: 32.

100 papers.

Sort:

By area:

  1. 2024Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviewscited by 160
  2. 2024Mapping the Increasing Use of LLMs in Scientific Paperscited by 112
  3. 2024Watermark Stealing in Large Language Modelscited by 79
  4. 2023Mark My Words: Analyzing and Evaluating Language Model Watermarkscited by 62
  5. 2024Adaptive Text Watermark for Large Language Modelscited by 60
  6. 2024Watermarking Makes Language Models Radioactivecited by 52
  7. 2024WaterMax: breaking the LLM watermark detectability-robustness-quality trade-offcited by 48
  8. 2023A Resilient and Accessible Distribution-Preserving Watermark for Large Language Modelscited by 47
  9. 2024No Free Lunch in LLM Watermarking: Trade-offs in Watermarking Design Choicescited by 47
  10. 2024SoK: Watermarking for AI-Generated Contentcited by 43
  11. 2024Copyright Protection in Generative AI: A Technical Perspectivecited by 35
  12. 2023Optimizing watermarks for large language modelscited by 22
38 more
  1. 2024Black-Box Detection of Language Model Watermarkscited by 22
  2. 2025AdaDetectGPT: Adaptive Detection of LLM-Generated Text with Statistical Guaranteescited by 20
  3. 2024Learning to Watermark LLM-generated Text via Reinforcement Learningcited by 19
  4. 2024Double-I Watermark: Protecting Model Copyright for LLM Fine-tuningcited by 18
  5. 2024Debiasing Watermarks for Large Language Models via Maximal Couplingcited by 18
  6. 2025BiMark: Unbiased Multilayer Watermarking for Large Language Modelscited by 18
  7. 2024Permute-and-Flip: An optimally stable and watermarkable decoder for LLMscited by 17
  8. 2024Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approachcited by 17
  9. 2023Necessary and Sufficient Watermark for Large Language Modelscited by 16
  10. 2024A Watermark for Order-Agnostic Language Modelscited by 14
  11. 2025Mark Your LLM: Detecting the Misuse of Open-Source Large Language Models via Watermarkingcited by 14
  12. 2024Edit Distance Robust Watermarks via Indexing Pseudorandom Codescited by 12
  13. 2024De-mark: Watermark Removal in Large Language Modelscited by 11
  14. 2025Scalable Fingerprinting of Large Language Modelscited by 11
  15. 2025Towards Watermarking of Open-Source LLMscited by 11
  16. 2025StealthInk: A Multi-bit and Stealthy Watermark for Large Language Modelscited by 11
  17. 2024Distortion-free Watermarks are not Truly Distortion-free under Watermark Key Collisionscited by 10
  18. 2024Watermarking Language Models with Error Correcting Codescited by 10
  19. 2024Can Watermarked LLMs be Identified by Users via Crafted Prompts?cited by 10
  20. 2025Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learningcited by 10
  21. 2025Unified attacks to large language model watermarks: spoofing and scrubbing in unauthorized knowledge distillationcited by 9
  22. 2025Optimized Couplings for Watermarking Large Language Modelscited by 9
  23. 2025In-Context Watermarks for Large Language Modelscited by 9
  24. 2024Inevitable Trade-off between Watermark Strength and Speculative Sampling Efficiency for Language Modelscited by 7
  25. 2025Character-Level Perturbations Disrupt LLM Watermarkscited by 7
  26. 2025Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoptioncited by 7
  27. 2024WaterPool: A Watermark Mitigating Trade-offs among Imperceptibility, Efficacy and Robustnesscited by 6
  28. 2024Efficiently Identifying Watermarked Segments in Mixed-Source Textscited by 6
  29. 2024SVIP: Towards Verifiable Inference of Open-source Large Language Modelscited by 6
  30. 2025SAEMark: Steering Personalized Multilingual LLM Watermarks with Sparse Autoencoderscited by 5
  31. 2025On the Empirical Power of Goodness-of-Fit Tests in Watermark Detectioncited by 5
  32. 2026SLICE: Semantic Latent Injection via Compartmentalized Embedding for Image Watermarkingcited by 5
  33. 2024On Evaluating The Performance of Watermarked Machine-Generated Texts Under Adversarial Attackscited by 4
  34. 2024Building Intelligence Identification System via Large Language Model Watermarking: A Survey and Beyondcited by 4
  35. 2024Discovering Spoofing Attempts on Language Model Watermarkscited by 4
  36. 2024Segmenting Watermarked Texts From Language Modelscited by 4
  37. 2025Can you Finetune your Binoculars? Embedding Text Watermarks into the Weights of Large Language Modelscited by 4
  38. 2025Robustness Assessment and Enhancement of Text Watermarking for Google's SynthIDcited by 4

50 most cited of 100 All 100 (JSON)