Addressing Tokenization Inconsistency in Steganography and Watermarking Based on Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yan, Ruiyi, Murawaki, Yugo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909757943578624
author Yan, Ruiyi
Murawaki, Yugo
author_facet Yan, Ruiyi
Murawaki, Yugo
contents Large language models have significantly enhanced the capacities and efficiency of text generation. On the one hand, they have improved the quality of text-based steganography. On the other hand, they have also underscored the importance of watermarking as a safeguard against malicious misuse. In this study, we focus on tokenization inconsistency (TI) between Alice and Bob in steganography and watermarking, where TI can undermine robustness. Our investigation reveals that the problematic tokens responsible for TI exhibit two key characteristics: infrequency and temporariness. Based on these findings, we propose two tailored solutions for TI elimination: a stepwise verification method for steganography and a post-hoc rollback method for watermarking. Experiments show that (1) compared to traditional disambiguation methods in steganography, directly addressing TI leads to improvements in fluency, imperceptibility, and anti-steganalysis capacity; (2) for watermarking, addressing TI enhances detectability and robustness against attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20718
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Addressing Tokenization Inconsistency in Steganography and Watermarking Based on Large Language Models
Yan, Ruiyi
Murawaki, Yugo
Computation and Language
Large language models have significantly enhanced the capacities and efficiency of text generation. On the one hand, they have improved the quality of text-based steganography. On the other hand, they have also underscored the importance of watermarking as a safeguard against malicious misuse. In this study, we focus on tokenization inconsistency (TI) between Alice and Bob in steganography and watermarking, where TI can undermine robustness. Our investigation reveals that the problematic tokens responsible for TI exhibit two key characteristics: infrequency and temporariness. Based on these findings, we propose two tailored solutions for TI elimination: a stepwise verification method for steganography and a post-hoc rollback method for watermarking. Experiments show that (1) compared to traditional disambiguation methods in steganography, directly addressing TI leads to improvements in fluency, imperceptibility, and anti-steganalysis capacity; (2) for watermarking, addressing TI enhances detectability and robustness against attacks.
title Addressing Tokenization Inconsistency in Steganography and Watermarking Based on Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2508.20718