Abstract:Recently, due to the considerable feature expression ability of neural networks, deep linguistic steganalysis methods have been greatly developed. However, there are still two issues that need to be ameliorated. First, the prevailing linguistic steganalysis methods rely heavily on massive training data, which is labor-intensive and time-consuming. Second, these methods implement steganalysis only in different weak-concealed scenarios, the stego texts in each of which have only a single language style and payload. But in practice, the intercepted network samples are probably the mixture of the stego texts that possess different language styles and payloads, in which the semantic spatial distribution may be more chaotic than that in weak-concealed scenarios, thus making steganalysis more difficult. To address the above issues, a novel linguistic steganalysis method is proposed in this letter. First, the pre-trained BERT language model is constructed as an embedder to compensate for the shortage of data. Then, in addition to learning local and global semantic features, a feature interaction module is designed for exploring mutual effects between them. Furthermore, besides the typical cross-entropy loss, triplet loss is also introduced for the model training. In this way, the proposed method can refine more comprehensive and discriminative deep features in the intricate semantic space. The performance of the proposed method is compared with the representative linguistic steganalysis methods on datasets of different scales, and the experimental results reveal the superiority of the proposed method.

Linguistic Steganalysis Toward Social Network

A New Steganalysis Approach Based on Both Complexity Estimate and Statistical Filter

State-of-the-art Advances of Deep-learning Linguistic Steganalysis Research

CATS: Connection-Aware and Interaction-Based Text Steganalysis in Social Networks

High-Performance Linguistic Steganalysis, Capacity Estimation and Steganographic Positioning.

Small-Scale Linguistic Steganalysis for Multi-Concealed Scenarios

Linguistic Steganalysis in Few-Shot Scenario

Hi-Stega: A Hierarchical Linguistic Steganography Framework Combining Retrieval and Generation

An Effective Linguistic Steganalysis Framework Based on Hierarchical Mutual Learning

LINK: Linguistic Steganalysis Framework with External Knowledge

SeSy: Linguistic Steganalysis Framework Integrating Semantic and Syntactic Features

Linguistic Steganalysis with Graph Neural Networks

Linguistic Steganalysis Via Densely Connected LSTM with Feature Pyramid

Linguistic Steganography: from Symbolic Space to Semantic Space

Linguistic Steganography Detection Using Statistical Characteristics of Correlations Between Words

TStego-THU: Large-Scale Text Steganalysis Dataset

TS-CNN: Text Steganalysis from Semantic Space Based on Convolutional Neural Network

Towards Next-Generation Steganalysis: LLMs Unleash the Power of Detecting Steganography

SCL-Stega: Exploring Advanced Objective in Linguistic Steganalysis using Contrastive Learning

Co-Stega: Collaborative Linguistic Steganography for the Low Capacity Challenge in Social Media

Steganalysis against substitution-based linguistic steganography based on context clusters