Machine Translation Weekly 58: Poisoning machine translation
Today, I am going to talk about a topic that is rather unknown to me: the
safety and vulnerability of machine translation. I will comment on a paper
Targeted Poisoning Attacks on Black-Box Neural Machine
Translation by authors from the
University of Melbourne and Facebook AI.
The main issue making machine-translation users vulnerable is that they
typically do not understand the target language and do not have any other
choice than trusting the system that target-language output is adequate. Most
machine translation systems are provided as web services, so it seems that the
only one that can manipulate what the system produces is the one who runs the
web service. This is to a large extent true but in this case, the Achilles heel
of the system is the