Machine Translation Weekly 65: Sequence-to-sequence models and substitution ciphers
Today, I am going to talk about a recent pre-print on sequence-to-sequence
models for deciphering substitution ciphers. Doing such a thing was somewhere
at the bottom of my todo list for a few years, I suggested it as a thesis topic
to several master students and no one wanted to do it, so I am glad that
someone finally did the experiments. The title of the preprint is Can
Sequence-to-Sequence Models Crack Substitution
Ciphers? and the authors are from the
University of Southern California.
They work with 1:1 substitution
ciphers. This means that
every character in the text gets systematically replaced by another one, but
always the same one. Such ciphers are easy to crack based on the frequency of
the characters. If you would do that manually,