A Content-Preserving Secure Linguistic Steganography

2025-11-16

AI Summary

Existing linguistic steganography methods rely on content rewriting, which introduces detectable statistical anomalies that compromise covertness. This paper proposes a novel **content-preserving linguistic steganography paradigm**: rather than altering any character in the original text, it embeds secret information solely by controllably modulating the output probability distribution of a masked language model (MLM). Our key contributions are: (1) a dynamic distribution encoding strategy coupled with an enhanced masking mechanism to precisely identify embedding positions; (2) fine-tuning of the MLM to construct a reversible steganographic model, enabling lossless mapping from plaintext to stegotext and achieving 100% accurate secret extraction; and (3) theoretical perfect secrecy, with empirical results demonstrating significant improvements over state-of-the-art methods in embedding capacity, text naturalness, and security.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Existing linguistic steganography methods primarily rely on content transformations to conceal secret messages. However, they often cause subtle yet looking-innocent deviations between normal and stego texts, posing potential security risks in real-world applications. To address this challenge, we propose a content-preserving linguistic steganography paradigm for perfectly secure covert communication without modifying the cover text. Based on this paradigm, we introduce CLstega ( extit{C}ontent-preserving extit{L}inguistic extit{stega}nography), a novel method that embeds secret messages through controllable distribution transformation. CLstega first applies an augmented masking strategy to locate and mask embedding positions, where MLM(masked language model)-predicted probability distributions are easily adjustable for transformation. Subsequently, a dynamic distribution steganographic coding strategy is designed to encode secret messages by deriving target distributions from the original probability distributions. To achieve this transformation, CLstega elaborately selects target words for embedding positions as labels to construct a masked sentence dataset, which is used to fine-tune the original MLM, producing a target MLM capable of directly extracting secret messages from the cover text. This approach ensures perfect security of secret messages while fully preserving the integrity of the original cover text. Experimental results show that CLstega can achieve a 100% extraction success rate, and outperforms existing methods in security, effectively balancing embedding capacity and security.

Problem

• Addresses security risks in linguistic steganography caused by text deviations
• Preserves cover text integrity while embedding secret messages securely
• Achieves perfect security through controllable distribution transformation methods

Innovation

• Content-preserving embedding without modifying cover text
• Controllable distribution transformation using masked language model
• Dynamic steganographic coding for perfect security and integrity