StealthInk: A Multi-bit and Stealthy Watermark for Large Language Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiang, Ya, Wu, Chuxiong, Boroujeny, Massieh Kordi, Mark, Brian, Zeng, Kai
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908395458527232
author Jiang, Ya
Wu, Chuxiong
Boroujeny, Massieh Kordi
Mark, Brian
Zeng, Kai
author_facet Jiang, Ya
Wu, Chuxiong
Boroujeny, Massieh Kordi
Mark, Brian
Zeng, Kai
contents Watermarking for large language models (LLMs) offers a promising approach to identifying AI-generated text. Existing approaches, however, either compromise the distribution of original generated text by LLMs or are limited to embedding zero-bit information that only allows for watermark detection but ignores identification. We present StealthInk, a stealthy multi-bit watermarking scheme that preserves the original text distribution while enabling the embedding of provenance data, such as userID, TimeStamp, and modelID, within LLM-generated text. This enhances fast traceability without requiring access to the language model's API or prompts. We derive a lower bound on the number of tokens necessary for watermark detection at a fixed equal error rate, which provides insights on how to enhance the capacity. Comprehensive empirical evaluations across diverse tasks highlight the stealthiness, detectability, and resilience of StealthInk, establishing it as an effective solution for LLM watermarking applications.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05502
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle StealthInk: A Multi-bit and Stealthy Watermark for Large Language Models
Jiang, Ya
Wu, Chuxiong
Boroujeny, Massieh Kordi
Mark, Brian
Zeng, Kai
Cryptography and Security
Artificial Intelligence
Watermarking for large language models (LLMs) offers a promising approach to identifying AI-generated text. Existing approaches, however, either compromise the distribution of original generated text by LLMs or are limited to embedding zero-bit information that only allows for watermark detection but ignores identification. We present StealthInk, a stealthy multi-bit watermarking scheme that preserves the original text distribution while enabling the embedding of provenance data, such as userID, TimeStamp, and modelID, within LLM-generated text. This enhances fast traceability without requiring access to the language model's API or prompts. We derive a lower bound on the number of tokens necessary for watermark detection at a fixed equal error rate, which provides insights on how to enhance the capacity. Comprehensive empirical evaluations across diverse tasks highlight the stealthiness, detectability, and resilience of StealthInk, establishing it as an effective solution for LLM watermarking applications.
title StealthInk: A Multi-bit and Stealthy Watermark for Large Language Models
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2506.05502