arXiv Computation and Language

Is Multilingual LLM Watermarking Truly Multilingual? Scaling Robustness to 100+ Languages via Back-Translation

arXiv Computation and Language
Sep 7

Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings

The paper introduces Dual-Embedding Watermarking (DEW), a semantic watermarking technique for large language models that combines contextual and token-level embeddings. DEW applies algebraic vector-space operations to generate a watermark signal that remains robust to paraphrasing and translation, while obfuscating the signal with pseudo-random matrices seeded by a secret key. Experiments demonstrate state‑of‑the‑art robustness, especially against translation, with minimal computational overhead and preserved text quality at lower watermark strengths.

By Jonas Sch\"afer, Cezary Pilaszewicz, Gerhard Wunder
Hugging Face Trending Papers
Aug 20

Auditing Cross-Lingual Fairness in Language Model Watermarking

Watermarking schemes for large language model output are evaluated almost exclusively on English text using each scheme's detection threshold and a narrow set of quality measurements. Multilingual deployment exposes evaluation-design choices that are inconsequential on English but determine conclusions cross-lingually.

arXiv AI
5d ago

Watermarking Diffusion Language Models

The paper introduces the first watermark designed specifically for diffusion language models (DLMs), which generate tokens in arbitrary order unlike traditional autoregressive models. It overcomes the challenge of missing prior tokens by applying the watermark in expectation over the context and promoting tokens that strengthen the watermark when used as context. Experiments show a >99% true positive rate with minimal quality loss and comparable robustness to existing autoregressive watermarks.

By Thibaud Gloaguen, Robin Staab, Nikola Jovanovi\'c, Martin Vechev
arXiv Machine Learning
Aug 21

Auditing Cross-Lingual Fairness in Language Model Watermarking

arXiv:2608. 20047v1 Announce Type: cross Abstract: Watermarking schemes for large language model output are evaluated almost exclusively on English text using each scheme's detection threshold and a narrow set of quality measurements.

By Alexander Nemecek, Osama Zafar, Debargha Ganguly, Vikash Singh, Vipin Chaudhary, Erman Ayday
arXiv Computation and Language
Sep 11

What Language is This? Ask Your Tokenizer

The paper introduces UniLID, a lightweight language identification method that uses the UnigramLM tokenization algorithm to predict a string’s language by evaluating which language’s unigram distribution best explains the text. UniLID is data‑ and compute‑efficient, allows incremental addition of new languages without retraining, and can be integrated into existing tokenization pipelines. Experiments show competitive performance against baselines such as fasttext, GlotLID‑M, and CLD3, achieving 69% accuracy with five labeled samples per language and 89% with 25, and delivering significant gains on fine‑grained dialect identification.

By Clara Meister, Ahmetcan Yavuz, Pietro Lesci, Tiago Pimentel