Arithmetic coding with GPT-2 produces cover text that matches the language model's own distribution (KL about 4e-8 nats) and is rated by humans as plausible, enabling near-exact-distribution linguistic steganography.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Neural Linguistic Steganography
Arithmetic coding with GPT-2 produces cover text that matches the language model's own distribution (KL about 4e-8 nats) and is rated by humans as plausible, enabling near-exact-distribution linguistic steganography.