# Linguistics-Aware Non-Distortionary LLM Watermarking

arXiv: https://arxiv.org/abs/2606.00613 · HTML: https://mlsanity.com/paper/2606.00613/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security.
- Builds on 14 of its 56 references that are in the map.
- First posted on arXiv 30 May 2026.

## Placed in

- LLM safety and security / [Watermarking generated content and models](https://mlsanity.com/safety/watermarking-generated-content-and-models/)

## Builds on (most cited first, 14 of 14)

- [XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages](https://mlsanity.com/paper/2106.13822/) (2021)
- [Robust Distortion-free Watermarks for Language Models](https://mlsanity.com/paper/2307.15593/) (2023)
- [CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages](https://mlsanity.com/paper/2309.09400/) (2023)
- [Who Wrote this Code? Watermarking for Code Generation](https://mlsanity.com/paper/2305.15060/) (2023)
- [MLSUM: The Multilingual Summarization Corpus](https://mlsanity.com/paper/2004.14900/) (2020)
- [An Entropy-based Text Watermarking Detection Method](https://mlsanity.com/paper/2403.13485/) (2024)
- [GumbelSoft: Diversified Language Model Watermarking via the GumbelMax-trick](https://mlsanity.com/paper/2402.12948/) (2024)
- [The Stem Cell Hypothesis: Dilemma behind Multi-Task Learning with Transformer Encoders](https://mlsanity.com/paper/2109.06939/) (2021)
- [Revisiting the Robustness of Watermarking to Paraphrasing Attacks](https://mlsanity.com/paper/2411.05277/) (2024)
- [Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach](https://mlsanity.com/paper/2410.02890/) (2024)
- [Does Incomplete Syntax Influence Korean Language Model? Focusing on Word Order and Case Markers](https://mlsanity.com/paper/2407.09184/) (2024)
- [Evaluating the Robustness and Accuracy of Text Watermarking Under Real-World Cross-Lingual Manipulations](https://mlsanity.com/paper/2502.16699/) (2025)
- [A Linguistics-Aware LLM Watermarking via Syntactic Predictability](https://mlsanity.com/paper/2510.13829/) (2025)
- [Is Multilingual LLM Watermarking Truly Multilingual? Scaling Robustness to 100+ Languages via Back-Translation](https://mlsanity.com/paper/2510.18019/) (2025)

## Abstract (arXiv)

Watermarking should identify language-model output without degrading quality or limiting verification to the model provider. Multilingual deployment makes this harder because morphology, segmentation, and script change where watermark evidence can be naturally embedded. We introduce LUNA, a linguistically adaptive watermark that combines model-free detection with single-token non-distortion under the standard random-key model. LUNA estimates normalized next-tag entropy from part-of-speech contexts in an external corpus and uses it to set the depth of a non-distortionary binary tournament sampler; the detector reconstructs the same schedule from text, a tokenizer, a tagger, and a secret key. We evaluate LUNA on six typologically diverse languages and two domains against eight primary baselines. LUNA attains an AUROC of 0.9959 and the lowest mean absolute median perplexity shift, 0.045, across the 12 settings; its 95% bootstrap interval [0.022, 0.073] lies below all baseline intervals. LUNA also records the lowest mean on Self-BLEU, Distinct-1, surprisal, and entropy shifts; it is the only method that simultaneously achieves AUROC > 0.99 and an absolute median perplexity shift below 0.1 in a majority of settings, reaching this regime in 9 of the 12 settings while no baseline reaches it in more than 2. Our code is available at https://github.com/Shinwoo-Park/luna_watermark.