mlsanityData to 7 Oct 2026

k-SemStamp: A Clustering-Based Semantic Watermark for Detection of Machine-Generated Text

arXiv 2402.11399 ↗First posted 17 Feb 2024cs.CL cs.CR cs.CYcited by 46

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

Abstract (arXiv)

Recent watermarked generation algorithms inject detectable signatures during language generation to facilitate post-hoc detection. While token-level watermarks are vulnerable to paraphrase attacks, SemStamp (Hou et al., 2023) applies watermark on the semantic representation of sentences and demonstrates promising robustness. SemStamp employs locality-sensitive hashing (LSH) to partition the semantic space with arbitrary hyperplanes, which results in a suboptimal tradeoff between robustness and speed. We propose k-SemStamp, a simple yet effective enhancement of SemStamp, utilizing k-means clustering as an alternative of LSH to partition the embedding space with awareness of inherent semantic structure. Experimental results indicate that k-SemStamp saliently improves its robustness and sampling efficiency while preserving the generation quality, advancing a more effective tool for machine-generated text detection.

Key facts. Cited by 46 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 14 of its 21 references that are in the map. First posted on arXiv 17 Feb 2024.

Builds on

14 of 21 references.

Closest ancestors · 6

Same direction, newest first

  1. 2023Robust Distortion-free Watermarks for Language Models28 Jul
  2. 2023Watermarking Conditional Text Generation for AI Detection: Unveiling Challenges and a Semantic-Aware Watermark Remedy25 Jul
  3. 2023Provable Robust Watermarking for AI-Generated Text30 Jun
  4. 2023Robust Multi-bit Natural Language Watermarking through Invariant Features3 May
  5. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense23 Mar
  6. 2023Can AI-Generated Text be Reliably Detected?17 Mar

Other references · 8

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  3. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  4. 2019PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarizationcited by 1,100
  5. 2018Model Cards for Model Reportingcited by 1,068
  6. 2021BookSum: A Collection of Datasets for Long-form Narrative Summarizationcited by 184
  7. 2018Generating Informative and Diverse Conversational Responses via Adversarial Information Maximizationcited by 114
  8. 2022Measuring and Improving Semantic Diversity of Dialogue Generationcited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 7; 2025: 20; 2026: 19.

46 papers.

Sort:

By area:

  1. 2024SoK: Watermarking for AI-Generated Contentcited by 43
  2. 2024Stumbling Blocks: Stress Testing the Robustness of Machine-Generated Text Detectors Under Attackscited by 25
  3. 2024PostMark: A Robust Blackbox Watermark for Large Language Modelscited by 24
  4. 2024Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Surveycited by 23
  5. 2024Watermarking Techniques for Large Language Models: A Surveycited by 20
  6. 2025CoheMark: A Novel Sentence-Level Watermark for Enhanced Text Qualitycited by 17
  7. 2025Can LLM Watermarks Robustly Prevent Unauthorized Knowledge Distillation?cited by 16
  8. 2025PMark: Towards Robust and Distortion-free Semantic-level Watermarking with Channel Constraintscited by 14
  9. 2025SimMark: A Robust Sentence-Level Similarity-Based Watermarking Algorithm for Large Language Modelscited by 13
  10. 2025DETree: DEtecting Human-AI Collaborative Texts via Tree-Structured Hierarchical Representation Learningcited by 12
  11. 2024Watermarking Language Models with Error Correcting Codescited by 10
  12. 2025Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learningcited by 10
34 more
  1. 2025Unified attacks to large language model watermarks: spoofing and scrubbing in unauthorized knowledge distillationcited by 9
  2. 2025Enhancing LLM Watermark Resilience Against Both Scrubbing and Spoofing Attackscited by 8
  3. 2024Topic-Based Watermarks for Large Language Modelscited by 7
  4. 2025From Trade-off to Synergy: A Versatile Symbiotic Watermarking Framework for Large Language Modelscited by 5
  5. 2025Invisible Entropy: Towards Safe and Efficient Low-Entropy LLM Watermarkingcited by 5
  6. 2025Large Language Model Sourcing: A Surveycited by 5
  7. 2025Evaluating the Robustness and Accuracy of Text Watermarking Under Real-World Cross-Lingual Manipulationscited by 4
  8. 2025Paladin: Defending LLM-enabled Phishing Emails with a New Trigger-Tag Paradigmcited by 4
  9. 2025WaterMod: Modular Token-Rank Partitioning for Probability-Balanced LLM Watermarkingcited by 4
  10. 2026PASA: A Principled Embedding-Space Watermarking Approach for LLM-Generated Text under Semantic-Invariant Attackscited by 4
  11. 2026Distilling the Thought, Watermarking the Answer: A Principle Semantic Guided Watermark for Large Reasoning Modelscited by 3
  12. 2025SoK: Are Watermarks in LLMs Ready for Deployment?cited by 2
  13. 2025SimKey: A Semantically Aware Key Module for Watermarking Language Modelscited by 2
  14. 2025How Good is Post-Hoc Watermarking With Language Model Rephrasing?cited by 2
  15. 2026TextSeal: A Localized LLM Watermark for Provenance & Distillation Protectioncited by 2
  16. 2026Robust LLM Watermarking with Minimal Semantic Distortion for IP Protectioncited by 1
  17. 2026SAMark: A Self-Anchored Text Watermarking with Paragraph-Level Paraphrase Robustnesscited by 1
  18. 2026Implicit Identity Technologies for LLMs: Fingerprinting and Watermarking across Datasets, Models, and Generated Contentcited by 1
  19. 2026WaterMoE: Expert-Routing-based Watermarking for High Fidelity and Efficiencycited by 1
  20. 2026DenMark: Robust Semantic Watermarking for Diffusion Language Modelscited by 1
  21. 2025Watermarking Large Language Models in Europe: Interpreting the AI Act in Light of Technologycited by 0
  22. 2025WaterSearch: Exploring Seed Pooling for Improving the Quality-Detectability Trade-off in LLM Watermarkingcited by 0
  23. 2025DualGuard: Dual-stream Large Language Model Watermarking Defense against Paraphrase and Spoofing Attackcited by 0
  24. 2026Semantic Differentiation for Tackling Challenges in Watermarking Low-Entropy Constrained Generation Outputscited by 0
  25. 2026WorldCup Sampling for Multi-bit LLM Watermarkingcited by 0
  26. 2026Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Frameworkcited by 0
  27. 2026LLM Output Detectability and Task Performance Can be Jointly Optimizedcited by 0
  28. 2026SWAN: Semantic Watermarking with Abstract Meaning Representationcited by 0
  29. 2026AliMark: Enhancing Robustness of Sentence-Level Watermarking Against Text Paraphrasingcited by 0
  30. 2026Robust Text Watermarking for Large Language Models via Dual Semantic Embeddingscited by 0
  31. 2026LLM Watermarking as Big Data Provenance: A Deployment-Oriented Systematizationcited by 0
  32. 2026Semantic Watermarking with Order-Robust Detection over Sub-sentence Unitscited by 0
  33. 2026OpenStamp: A Watermark for Open-Source Language Modelscited by 0
  34. 2026Beyond Semantic Narrowing: Robust and Efficient LLM Watermarking with Hamming Neighborhoodscited by 0