mlsanityData to 7 Oct 2026

LLM Watermarking Using Mixtures and Statistical-to-Computational Gaps

arXiv 2505.01484 ↗First posted 2 May 2025cs.CR cs.LGcited by 1

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

Abstract (arXiv)

Given a text, can we determine whether it was generated by a large language model (LLM) or by a human? A widely studied approach to this problem is watermarking. We propose an undetectable and elementary watermarking scheme in the closed setting. Also, in the harder open setting, where the adversary has access to most of the model, we propose an unremovable watermarking scheme.

Key facts. Cited by 1 paper 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 25 references that are in the map. First posted on arXiv 2 May 2025.

Builds on

14 of 25 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025GaussMark: A Practical Approach for Structural Watermarking of Language Models17 Jan
  2. 2024Provably Robust Watermarks for Open-Source Language Models24 Oct
  3. 2024A Statistical Framework of Watermarks for Large Language Models: Pivot, Detection Efficiency and Optimal Rules1 Apr
  4. 2023Robust Distortion-free Watermarks for Language Models28 Jul
  5. 2023Undetectable Watermarks for Language Models25 May
  6. 2023A Watermark for Large Language Models24 Jan

Other references · 8

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 3,972
  4. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  5. 2023The Curse of Recursion: Training on Generated Data Makes Models Forgetcited by 343
  6. 2020Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hidingcited by 121
  7. 2019Notes on Computational Hardness of Hypothesis Testing: Predictions using the Low-Degree Likelihood Ratiocited by 82
  8. 2020Reducibility and Statistical-Computational Gaps from Secret Leakagecited by 50

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Watermark Forensics for Generative Models: An Information-Theoretic Perspectivecited by 0