Saved in:
Bibliographic Details
Main Authors: Kim, Joeun, Kim, HoEun, Jin, Dongsup, Kim, Young-Sik
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.00348
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918477174931456
author Kim, Joeun
Kim, HoEun
Jin, Dongsup
Kim, Young-Sik
author_facet Kim, Joeun
Kim, HoEun
Jin, Dongsup
Kim, Young-Sik
contents Recent multi-bit watermarking methods for large language models (LLMs) prioritize capacity over reliability, often conflating decoding with detection. Our analysis reveals that existing ECC-based extractors suffer from catastrophic false positive rates (FPR), and applying rejection thresholds merely collapses detection sensitivity (TPR) to random guessing. To resolve this structural limitation, we propose \textbf{BREW} (Block-wise Reliable Embedding for Watermarking), a framework shifting the paradigm to \emph{designated verification}. BREW employs a two-stage mechanism: (i) \textbf{blind message estimation} via independent block voting, followed by (ii) \textbf{window-shifting verification} that rigorously validates the payload against local edits. Experiments demonstrate that BREW achieves a TPR of 0.965 with an FPR of 0.02 under 10\% synonym substitution, demonstrating that the high-FPR issue is not an inherent trade-off of multi-bit watermarking, but a solvable structural flaw of prior decoding-centric designs. Our framework is model-agnostic and theoretically grounded, providing a scalable solution for reliable forensic deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2605_00348
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Block-wise Codeword Embedding for Reliable Multi-bit Text Watermarking
Kim, Joeun
Kim, HoEun
Jin, Dongsup
Kim, Young-Sik
Cryptography and Security
Computation and Language
Recent multi-bit watermarking methods for large language models (LLMs) prioritize capacity over reliability, often conflating decoding with detection. Our analysis reveals that existing ECC-based extractors suffer from catastrophic false positive rates (FPR), and applying rejection thresholds merely collapses detection sensitivity (TPR) to random guessing. To resolve this structural limitation, we propose \textbf{BREW} (Block-wise Reliable Embedding for Watermarking), a framework shifting the paradigm to \emph{designated verification}. BREW employs a two-stage mechanism: (i) \textbf{blind message estimation} via independent block voting, followed by (ii) \textbf{window-shifting verification} that rigorously validates the payload against local edits. Experiments demonstrate that BREW achieves a TPR of 0.965 with an FPR of 0.02 under 10\% synonym substitution, demonstrating that the high-FPR issue is not an inherent trade-off of multi-bit watermarking, but a solvable structural flaw of prior decoding-centric designs. Our framework is model-agnostic and theoretically grounded, providing a scalable solution for reliable forensic deployment.
title Block-wise Codeword Embedding for Reliable Multi-bit Text Watermarking
topic Cryptography and Security
Computation and Language
url https://arxiv.org/abs/2605.00348