In plain words: Hiding secret messages in fluent text usually relies on assumed patterns in how natural text behaves. Testing those assumptions shows when earlier schemes can be spotted, and a new encoding gives stronger guarantees that the message stays unnoticed.
Abstract · Towards Near-imperceptible Steganographic Text
We show that the imperceptibility of several existing linguistic steganographic systems (Fang et al., 2017; Yang et al., 2018) relies on implicit assumptions on statistical behaviors of fluent text. We formally analyze them and empirically evaluate these assumptions. Furthermore, based on these observations, we propose an encoding algorithm called patient-Huffman with improved near-imperceptible guarantees.
Falcon Z. Dai, Zheng Cai
arXiv:1907.06679 · cs.CL, cs.CR, cs.LG · submitted Jul 15, 2019 · updated Jul 29, 2019
abstract · pdf · html · To appear at ACL 2019. Code available
