Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA)

Chen, Yingquan; Li, Qianmu; Bhattacharjya, Aniruddha; Wu, Xiaocong; Li, Huifeng; Chang, Qing; Zhu, Le; Xiao, Yan

doi:10.3390/app15179663

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA)

by

Yingquan Chen

¹

,

Qianmu Li

^1,2,*

,

Aniruddha Bhattacharjya

^3,*,†

,

Xiaocong Wu

⁴,

Huifeng Li

¹,

Qing Chang

⁴,

Le Zhu

⁵ and

Yan Xiao

⁶

¹

School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China

²

School of Cybersecurity, Tarim University, Alar City 843300, China

³

Department of Electronic Engineering, Tsinghua University, Beijing 100084, China

⁴

School of Cyber Science and Engineering, Nanjing University of Science and Technology, Wuxi 214443, China

⁵

Zhongke Yungang Technology Co., Ltd., Beijing 100010, China

⁶

Beijing Lianan Hengda Technology Co., Ltd., Beijing 100141, China

^*

Authors to whom correspondence should be addressed.

^†

PhD Alumni, Department of Electronic Engineering, Tsinghua University, Beijing 100190, China.

Appl. Sci. 2025, 15(17), 9663; https://doi.org/10.3390/app15179663

Submission received: 30 July 2025 / Revised: 28 August 2025 / Accepted: 29 August 2025 / Published: 2 September 2025

(This article belongs to the Special Issue Cyber Security and Software Engineering)

Download Versions Notes

Abstract

Generative linguistic steganography aims to produce texts that remain both perceptually and statistically imperceptible. The existing embedding algorithms often suffer from imbalanced candidate selection, where high-probability words are overlooked and low-probability words dominate, leading to reduced coherence and fluency. We introduce a character-based diffusion embedding algorithm (CDEA) that uniquely leverages character-level statistics and a power-law-inspired grouping strategy to better balance candidate word selection. Unlike prior methods, the proposed CDEA explicitly prioritizes high-probability candidates, thereby improving both semantic consistency and text naturalness. When combined with XLNet, it effectively generates longer sensitive sequences while preserving quality. The experimental results showed that CDEA not only produces steganographic texts with higher imperceptibility and fluency but also achieves stronger resistance to steganalysis compared with the existing approaches. Future work will be to enhance statistical imperceptibility, integrate CDEA with larger language models such as GPT-5, and extend applications to cross-lingual, multimodal, and practical IoT or blockchain communication scenarios.

Keywords: Generative linguistic steganography; embedding algorithm; character-based diffusion embedding algorithm (CDEA); perceptual imperceptibility

Share and Cite

MDPI and ACS Style

Chen, Y.; Li, Q.; Bhattacharjya, A.; Wu, X.; Li, H.; Chang, Q.; Zhu, L.; Xiao, Y. Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA). Appl. Sci. 2025, 15, 9663. https://doi.org/10.3390/app15179663

AMA Style

Chen Y, Li Q, Bhattacharjya A, Wu X, Li H, Chang Q, Zhu L, Xiao Y. Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA). Applied Sciences. 2025; 15(17):9663. https://doi.org/10.3390/app15179663

Chicago/Turabian Style

Chen, Yingquan, Qianmu Li, Aniruddha Bhattacharjya, Xiaocong Wu, Huifeng Li, Qing Chang, Le Zhu, and Yan Xiao. 2025. "Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA)" Applied Sciences 15, no. 17: 9663. https://doi.org/10.3390/app15179663

APA Style

Chen, Y., Li, Q., Bhattacharjya, A., Wu, X., Li, H., Chang, Q., Zhu, L., & Xiao, Y. (2025). Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA). Applied Sciences, 15(17), 9663. https://doi.org/10.3390/app15179663

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Enhancement of the Generation Quality of Generative Linguistic Steganographic Texts by a Character-Based Diffusion Embedding Algorithm (CDEA)

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI