mlsanityData to 7 Oct 2026

An Entropy-based Text Watermarking Detection Method

arXiv 2403.13485 ↗First posted 20 Mar 2024cs.CLcited by 61

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

Abstract (arXiv)

Text watermarking algorithms for large language models (LLMs) can effectively identify machine-generated texts by embedding and detecting hidden features in the text. Although the current text watermarking algorithms perform well in most high-entropy scenarios, its performance in low-entropy scenarios still needs to be improved. In this work, we opine that the influence of token entropy should be fully considered in the watermark detection process, i.e., the weight of each token during watermark detection should be customized according to its entropy, rather than setting the weights of all tokens to the same value as in previous methods. Specifically, we propose Entropy-based Text Watermarking Detection (EWD) that gives higher-entropy tokens higher influence weights during watermark detection, so as to better reflect the degree of watermarking. Furthermore, the proposed detection process is training-free and fully automated. From the experiments, we demonstrate that our EWD can achieve better detection performance in low-entropy scenarios, and our method is also general and can be applied to texts with different entropy distributions. Our code and data is availablefootnoteurlhttps://github.com/luyijian3/EWD. Additionally, our algorithm could be accessed through MarkLLM citepan2024markllmfootnoteurlhttps://github.com/THU-BPM/MarkLLM.

Key facts. Cited by 61 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 15 of its 30 references that are in the map. First posted on arXiv 20 Mar 2024.

Builds on

15 of 30 references.

Closest ancestors · 8

Same direction, newest first

  1. 2023A Survey of Text Watermarking in the Era of Large Language Models13 Dec
  2. 2023A Robust Semantics-based Watermark for Large Language Model against Paraphrasing15 Nov
  3. 2023Publicly-Detectable Watermarking for Language Models27 Oct
  4. 2023Robust Distortion-free Watermarks for Language Models28 Jul
  5. 2023Provable Robust Watermarking for AI-Generated Text30 Jun
  6. 2023Who Wrote this Code? Watermarking for Code Generation24 May
  7. 2023Watermarking Text Generated by Black-Box Language Models14 May
  8. 2023Robust Multi-bit Natural Language Watermarking through Invariant Features3 May

Other references · 7

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2021Program Synthesis with Large Language Modelscited by 3,509
  4. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  5. 2023StarCoder: may the source be with you!cited by 790
  6. 2017Challenges in Data-to-Document Generationcited by 226
  7. 2021Text-to-Table: A New Way of Information Extractioncited by 47

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 8; 2025: 35; 2026: 18.

61 papers.

Sort:

By area:

  1. 2024MarkLLM: An Open-Source Toolkit for LLM Watermarkingcited by 80
  2. 2025Towards Trustworthy Retrieval Augmented Generation for Large Language Models: A Surveycited by 29
  3. 2024Revisiting the Robustness of Watermarking to Paraphrasing Attackscited by 19
  4. 2025CoheMark: A Novel Sentence-Level Watermark for Enhanced Text Qualitycited by 17
  5. 2025Can LLM Watermarks Robustly Prevent Unauthorized Knowledge Distillation?cited by 16
  6. 2025Mark Your LLM: Detecting the Misuse of Open-Source Large Language Models via Watermarkingcited by 14
  7. 2025PMark: Towards Robust and Distortion-free Semantic-level Watermarking with Channel Constraintscited by 14
  8. 2025Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attackscited by 12
  9. 2025VLA-Mark: A cross modal watermark for large vision-language alignment modelcited by 11
  10. 2024Can Watermarked LLMs be Identified by Users via Crafted Prompts?cited by 10
  11. 2024WaterSeeker: Pioneering Efficient Detection of Watermarked Segments in Large Documentscited by 9
  12. 2025Enhancing LLM Watermark Resilience Against Both Scrubbing and Spoofing Attackscited by 8
38 more
  1. 2024Watermark under Fire: A Robustness Evaluation of LLM Watermarkingcited by 7
  2. 2025Marking Code Without Breaking It: Code Watermarking for Detecting LLM-Generated Codecited by 7
  3. 2025From Trade-off to Synergy: A Versatile Symbiotic Watermarking Framework for Large Language Modelscited by 5
  4. 2025Invisible Entropy: Towards Safe and Efficient Low-Entropy LLM Watermarkingcited by 5
  5. 2025SAEMark: Steering Personalized Multilingual LLM Watermarks with Sparse Autoencoderscited by 5
  6. 2024On Evaluating The Performance of Watermarked Machine-Generated Texts Under Adversarial Attackscited by 4
  7. 2024Discovering Spoofing Attempts on Language Model Watermarkscited by 4
  8. 2025Evaluating the Robustness and Accuracy of Text Watermarking Under Real-World Cross-Lingual Manipulationscited by 4
  9. 2025Paladin: Defending LLM-enabled Phishing Emails with a New Trigger-Tag Paradigmcited by 4
  10. 2025DITTO: A Spoofing Attack Framework on Watermarked LLMs via Knowledge Distillationcited by 4
  11. 2025BiMarker: Enhancing Text Watermark Detection for Large Language Models with Bipolar Watermarkscited by 3
  12. 2025Watermarking Degrades Alignment in Language Models: Analysis and Mitigationcited by 3
  13. 2025Addressing Tokenization Inconsistency in Steganography and Watermarking Based on Large Language Modelscited by 3
  14. 2026Distilling the Thought, Watermarking the Answer: A Principle Semantic Guided Watermark for Large Reasoning Modelscited by 3
  15. 2026A Visual Semantic Adaptive Watermark grounded by Prefix-Tuning for Large Vision-Language Modelcited by 3
  16. 2025SoK: Are Watermarks in LLMs Ready for Deployment?cited by 2
  17. 2025EMMM, Explain Me My Model! Explainable Machine Generated Text Detection in Dialoguescited by 2
  18. 2025CATMark: A Context-Aware Thresholding Framework for Robust Cross-Task Watermarking in Large Language Modelscited by 2
  19. 2025A Linguistics-Aware LLM Watermarking via Syntactic Predictabilitycited by 2
  20. 2026Watermarking Should Be Treated as a Monitoring Primitivecited by 2
  21. 2025SPADE: Structured Prompting Augmentation for Dialogue Enhancement in Machine-Generated Text Detectioncited by 1
  22. 2025HeavyWater and SimplexWater: Distortion-Free LLM Watermarks for Low-Entropy Next-Token Predictionscited by 1
  23. 2025Learning to Watermark: A Selective Watermarking Framework for Large Language Models via Multi-Objective Optimizationcited by 1
  24. 2025Is Multilingual LLM Watermarking Truly Multilingual? Scaling Robustness to 100+ Languages via Back-Translationcited by 1
  25. 2026SearchLLM: Detecting LLM Paraphrased Text by Measuring the Similarity with Regeneration of the Candidate Source via Search Enginecited by 1
  26. 2026SAMark: A Self-Anchored Text Watermarking with Paragraph-Level Paraphrase Robustnesscited by 1
  27. 2026WaterMoE: Expert-Routing-based Watermarking for High Fidelity and Efficiencycited by 1
  28. 2024Watermark Smoothing Attacks against Language Modelscited by 0
  29. 2025Toward Copyright Integrity and Verifiability via Multi-Bit Watermarking for Intelligent Transportation Systemscited by 0
  30. 2025CEFW: A Comprehensive Evaluation Framework for Watermark in Large Language Modelscited by 0
  31. 2025Protecting Creative Writing Copyright against AI Imitation via Implicit Watermarkingcited by 0
  32. 2025Yet Another Watermark for Large Language Modelscited by 0
  33. 2025LLM Watermark Evasion via Bias Inversioncited by 0
  34. 2025Signature vs. Substance: Evaluating the Balance of Adversarial Resistance and Linguistic Quality in Watermarking Large Language Modelscited by 0
  35. 2025WaterSearch: Exploring Seed Pooling for Improving the Quality-Detectability Trade-off in LLM Watermarkingcited by 0
  36. 2025CODE ACROSTIC: Robust Watermarking for Code Generationcited by 0
  37. 2025DualGuard: Dual-stream Large Language Model Watermarking Defense against Paraphrase and Spoofing Attackcited by 0
  38. 2026Semantic Differentiation for Tackling Challenges in Watermarking Low-Entropy Constrained Generation Outputscited by 0

50 most cited of 61 All 61 (JSON)