Generative Steganography Based on Long Readable Text Generation

Dados Bibliográficos

ID22106941
AutoresYi Cao (0000-0001-9197-0763, Nanjing University of Information Science and Technology), Zhili Zhou (0000-0002-7515-507X, Guangzhou University), Chinmay Chakraborty (0000-0002-4385-0975, Birla Institute of Technology, Mesra), Meimin Wang (0000-0001-9313-5593, Nanjing University of Information Science and Technology), Q M Jonathan Wu (0000-0002-5208-7975, University of Windsor), Xingming Sun (Nanjing University of Information Science and Technology), Keping Yu (0000-0001-5735-2507, Hosei University)
Ano2024
Volume11
Fascículo4
Páginas4584-4594
Data de publicação2024-08-01
Peer ReviewedSim
Open AccessSim
TipoARTICLE
PeriódicoIEEE Transactions on Computational Social Systems (JOURNAL)
Identificadores do periódicoISSN: 2329-924X • E-ISSN: 2373-7476
EditoraInstitute of Electrical and Electronics Engineers (IEEE) (PUBLISHER)
DOI10.1109/tcss.2022.3174013
OpenAlexW4285213468
IdiomaEN
Citações recebidas3
Referências citadas49

Text steganography has received a lot of attention in the application of covert communication. How to ensure desirable capacity and imperceptibility has become a key issue in text steganography. There are two typical approaches, i.e., text-selection-based steganography and text-generation-based steganography. However, the text-selection-based approaches generally have the very low hidden capacity and are not applicable in practical scenarios. Although the text-generation-based approaches can embed secret messages with higher capacity during text generation, they are prone to semantic incoherence and semantic errors when generating long texts. To address the abovementioned issues, this article proposes a novel text steganography based on long readable text generation. It first determines the topic of the stego-text according to the scenarios of the communication parties. Then, the plug and play language model (PPLM) is explored to generate the long readable stego-text conforming to the topic with semantic coherency. A given secret message is hidden during text generation by selecting proper words in an established embeddable candidate word pool (ECWP). Establishing the ECWP prevents the language model (LM) from selecting words with low probability in the text generation, thereby avoiding the generation of low-quality or even grammatically incorrect stego-text. Experimental results show that the proposed approach significantly increases hidden capacity while maintaining good imperceptibility compared with the existing approaches

Computer security · Generative grammar · Information retrieval · Linguistics · Natural language processing · Speech recognition · Steganography · Text generation · Advanced Data and IoT Technologies · Advanced Steganography and Watermarking Techniques · Computer Science · Internet Traffic Analysis and Secure E-voting · Artificial Intelligence