mlsanityData to 7 Oct 2026

Watermarking Techniques for Large Language Models: A Survey

arXiv 2409.00089 ↗First posted 26 Aug 2024cs.CR cs.AIcited by 20

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

Abstract (arXiv)

With the rapid advancement and extensive application of artificial intelligence technology, large language models (LLMs) are extensively used to enhance production, creativity, learning, and work efficiency across various domains. However, the abuse of LLMs also poses potential harm to human society, such as intellectual property rights issues, academic misconduct, false content, and hallucinations. Relevant research has proposed the use of LLM watermarking to achieve IP protection for LLMs and traceability of multimedia data output by LLMs. To our knowledge, this is the first thorough review that investigates and analyzes LLM watermarking technology in detail. This review begins by recounting the history of traditional watermarking technology, then analyzes the current state of LLM watermarking research, and thoroughly examines the inheritance and relevance of these techniques. By analyzing their inheritance and relevance, this review can provide research with ideas for applying traditional digital watermarking techniques to LLM watermarking, to promote the cross-integration and innovation of watermarking technology. In addition, this review examines the pros and cons of LLM watermarking. Considering the current multimodal development trend of LLMs, it provides a detailed analysis of emerging multimodal LLM watermarking, such as visual and audio data, to offer more reference ideas for relevant research. This review delves into the challenges and future prospects of current watermarking technologies, offering valuable insights for future LLM watermarking research and applications.

Key facts. Cited by 20 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 95 of its 274 references that are in the map. First posted on arXiv 26 Aug 2024.

Builds on

95 of 274 references.

Closest ancestors · 39

Same direction, newest first

  1. 2024From Intentions to Techniques: A Comprehensive Taxonomy and Challenges in Text Watermarking for Large Language Models17 Jun
  2. 2024MarkLLM: An Open-Source Toolkit for LLM Watermarking16 May
  3. 2024Towards Better Statistical Understanding of Watermarking LLMs19 Mar
  4. 2024Duwak: Dual Watermarks in Large Language Models12 Mar
  5. 2024WaterMax: breaking the LLM watermark detectability-robustness-quality trade-off6 Mar
  6. 2024Token-Specific Watermarking with Enhanced Detectability and Semantic Coherence for Large Language Models28 Feb
  7. 2024EmMark: Robust Watermarks for IP Protection of Embedded Quantized Large Language Models27 Feb
  8. 2024Multi-Bit Distortion-Free Watermarking for Large Language Models26 Feb
  9. 2024Can Watermarks Survive Translation? On the Cross-lingual Consistency of Text Watermark for Large Language Models21 Feb
  10. 2024GumbelSoft: Diversified Language Model Watermarking via the GumbelMax-trick20 Feb
  11. 2024k-SemStamp: A Clustering-Based Semantic Watermark for Detection of Machine-Generated Text17 Feb
  12. 2024Proving membership in LLM pretraining data via data watermarks16 Feb
27 more
  1. 2023Optimizing watermarks for large language models28 Dec
  2. 2023Watermarking Vision-Language Pre-trained Models for Multi-modal Embedding as a Service10 Nov
  3. 2023Publicly-Detectable Watermarking for Language Models27 Oct
  4. 2023Watermarking LLMs with Weight Quantization17 Oct
  5. 2023Embarrassingly Simple Text Watermarks13 Oct
  6. 2023Invisible Watermarking for Audio Generation Diffusion Models22 Sep
  7. 2023Catch You Everything Everywhere: Guarding Textual Inversion via Concept Watermarking12 Sep
  8. 2023T2IW: Joint Text to Image & Watermark Generation7 Sep
  9. 2023Advancing Beyond Identification: Multi-bit Watermark for Large Language Models1 Aug
  10. 2023Watermarking Conditional Text Generation for AI Detection: Unveiling Challenges and a Semantic-Aware Watermark Remedy25 Jul
  11. 2023Provable Robust Watermarking for AI-Generated Text30 Jun
  12. 2023On the Reliability of Watermarks for Large Language Models7 Jun
  13. 2023DiffusionShield: A Watermark for Copyright Protection against Generative Diffusion Models25 May
  14. 2023Who Wrote this Code? Watermarking for Code Generation24 May
  15. 2023Watermarking Text Data on Large Language Models for Dataset Copyright22 May
  16. 2023Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermark17 May
  17. 2023Watermarking Text Generated by Black-Box Language Models14 May
  18. 2023PTW: Pivotal Tuning Watermarking for Pre-Trained Image Generators14 Apr
  19. 2023The Stable Signature: Rooting Watermarks in Latent Diffusion Models27 Mar
  20. 2023Protecting Language Generation Models via Invisible Watermarking6 Feb
  21. 2022Neural network fragile watermarking with no model performance degradation16 Aug
  22. 2022DynaMarks: Defending Against Deep Learning Model Extraction Using Dynamic Watermarking27 Jul
  23. 2022Cosine Model Watermarking Against Ensemble Distillation5 Mar
  24. 2022Attributable-Watermarking of Speech Generative Models17 Feb
  25. 2021Protecting the Intellectual Properties of Deep Neural Networks with an Additional Class and Steganographic Images19 Apr
  26. 2021Deep Model Intellectual Property Protection via Deep Watermarking8 Mar
  27. 2021Don't Forget to Sign the Gradients!5 Mar

Other references · 56

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  6. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
  9. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,462
  10. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  11. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  12. 2019SciBERT: A Pretrained Language Model for Scientific Textcited by 1,404
38 more
  1. 2017Efficient Processing of Deep Neural Networks: A Tutorial and Surveycited by 427
  2. 2023A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Uglycited by 380
  3. 2022A Systematic Evaluation of Large Language Models of Codecited by 360
  4. 2021Pre-Trained Models: Past, Present and Futurecited by 319
  5. 2017Embedding Watermarks into Deep Neural Networkscited by 227
  6. 2020Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Datacited by 200
  7. 2023Universal Self-Consistency for Large Language Model Generationcited by 198
  8. 2016A Theory of Generative ConvNetcited by 174
  9. 2023Summary of ChatGPT-Related Research and Perspective Towards the Future of Large Language Modelscited by 152
  10. 2023Large Language Models for Robotics: A Surveycited by 132
  11. 2023Large Language Models in Law: A Surveycited by 128
  12. 2020Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hidingcited by 121
  13. 2021Towards artificial general intelligence via a multimodal foundation modelcited by 112
  14. 2023SciEval: A Multi-Level Large Language Model Evaluation Benchmark for Scientific Researchcited by 112
  15. 2017Adversarial Frontier Stitching for Remote Neural Network Watermarkingcited by 110
  16. 2019DAWN: Dynamic Adversarial Watermarking of Neural Networkscited by 94
  17. 2021A survey of deep neural network watermarking techniquescited by 69
  18. 2017A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothingcited by 60
  19. 2023AI-Generated Content (AIGC): A Surveycited by 59
  20. 2023Large Language Models in Education: Vision and Opportunitiescited by 54
  21. 2019Generating Sentiment-Preserving Fake Online Reviews Using Neural Language Models and Their Human- and Machine-based Detectioncited by 50
  22. 2020Model Watermarking for Image Processing Networkscited by 49
  23. 2020A Systematic Review on Model Watermarking for Neural Networkscited by 49
  24. 2023Learning to Generate Better Than Your LLMcited by 49
  25. 2019Robust Watermarking of Neural Network with Exponential Weightingcited by 48
  26. 2021Protecting Intellectual Property of Generative Adversarial Networks from Ambiguity Attackcited by 45
  27. 2023WavMark: Watermarking for Audio Generationcited by 41
  28. 2024Large Language Models for Medicine: A Surveycited by 38
  29. 2024Large Language Models for Education: A Surveycited by 32
  30. 2018A Survey of Utility-Oriented Pattern Miningcited by 27
  31. 2023Model-as-a-Service (MaaS): A Surveycited by 24
  32. 2021Traceability Transformed: Generating more Accurate Links with Pre-Trained BERT Modelscited by 20
  33. 2021Reversible Watermarking in Deep Convolutional Neural Networks for Integrity Authenticationcited by 17
  34. 2023Functional Invariants to Watermark Large Transformerscited by 14
  35. 2021Structural Watermarking to Deep Neural Networks via Network Channel Pruningcited by 10
  36. 2024WARDEN: Multi-Directional Backdoor Watermarks for Embedding-as-a-Service Copyright Protectioncited by 9
  37. 2022Metaverse Security and Privacy: An Overviewcited by 8
  38. 2021A Compact Neural Network-based Algorithm for Robust Image Watermarkingcited by 6

50 most cited of 56 All 56 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 11; 2026: 9.

20 papers.

Sort:

By area:

  1. 2025DOGe: Defensive Output Generation for LLM Protection Against Knowledge Distillationcited by 14
  2. 2025SoK: Large Language Model Copyright Auditing via Fingerprintingcited by 14
  3. 2025VLA-Mark: A cross modal watermark for large vision-language alignment modelcited by 11
  4. 2025Diversity Boosts AI-Generated Text Detectioncited by 6
  5. 2026Towards Distillation-Resistant Large Language Models: An Information-Theoretic Perspectivecited by 5
  6. 2025Video Signature: Implicit Watermarking for Video Diffusion Modelscited by 3
  7. 2026Distilling the Thought, Watermarking the Answer: A Principle Semantic Guided Watermark for Large Reasoning Modelscited by 3
  8. 2026PeerCheck: Enhancing LLM-Generated Academic Reviews Towards Human-Level Qualitycited by 3
  9. 2025SoK: Are Watermarks in LLMs Ready for Deployment?cited by 2
  10. 2025CATMark: A Context-Aware Thresholding Framework for Robust Cross-Task Watermarking in Large Language Modelscited by 2
  11. 2026Echoes within the Reasoning: Stealthy and Effective Watermarking via Chain of Thoughtcited by 2
  12. 2025Detection Avoidance Techniques for Large Language Modelscited by 1
8 more
  1. 2025LexiMark: Robust Watermarking via Lexical Substitutions to Enhance Membership Verification of an LLM's Textual Training Datacited by 1
  2. 2026Implicit Identity Technologies for LLMs: Fingerprinting and Watermarking across Datasets, Models, and Generated Contentcited by 1
  3. 2026WaterMoE: Expert-Routing-based Watermarking for High Fidelity and Efficiencycited by 1
  4. 2025An LLM Framework For Cryptography Over Chat Channelscited by 0
  5. 2025Watermarking Large Language Models in Europe: Interpreting the AI Act in Light of Technologycited by 0
  6. 2026Combating Data Laundering in LLM Trainingcited by 0
  7. 2026OpenStamp: A Watermark for Open-Source Language Modelscited by 0
  8. 2026Beyond Semantic Narrowing: Robust and Efficient LLM Watermarking with Hamming Neighborhoodscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Intrinsic Fingerprint of LLMs: Continue Training is NOT All You Need to Steal A Model!
  2. 2025Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoption
  3. 2025RAG-WM: An Efficient Black-Box Watermarking Approach for Retrieval-Augmented Generation of Large Language Models
  4. 2025SoK: The Privacy Paradox of Large Language Models: Advancements, Privacy Risks, and Mitigation
  5. 2025Security and Detectability Analysis of Unicode Text Watermarking Methods against Large Language Models
  6. 2024Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Survey
  7. 2023Source Attribution for Large Language Model-Generated Data
  8. 2023WatME: Towards Lossless Watermarking Through Lexical Redundancy