Generating Steganographic Text with LSTMs

Abstract

Motivated by concerns for user privacy, we design a steganographic system ("stegosystem") that enables two users to exchange encrypted messages without an adversary detecting that such an exchange is taking place. We propose a new linguistic stegosystem based on a Long Short-Term Memory (LSTM) neural network. We demonstrate our approach on the Twitter and Enron email datasets and show that it yields high-quality steganographic text while significantly improving capacity (encrypted bits per word) relative to the state-of-the-art.

ACL 2017 Student Research Workshop

References in corpus (1)

• Exploring the Limits of Language Modeling

Cited by in corpus (9)

• Automatically Generate Steganographic Text Based on Markov Model and Huffman Coding
• TS-CNN: Text Steganalysis from Semantic Space Based on Convolutional Neural Network
• AAG-Stega: Automatic Audio Generation-based Steganography
• Graph-Stega: Semantic Controllable Steganographic Text Generation Guided by Knowledge Graph
• Real-Time Steganalysis for Stream Media Based on Multi-channel Convolutional Sliding Windows
• HiDDeN: Hiding Data With Deep Networks
• Neural Linguistic Steganography
• Text Steganalysis with Attentional LSTM-CNN
• Generating Watermarked Adversarial Texts