Week 1 quiz had a question about the task of guessing a masked token by building bidirectional representations. I don’t recall seeing this in any of the lectures up to this point or in the 2017 paper. Where was this covered?
Week 1 quiz had a question about the task of guessing a masked token by building bidirectional representations. I don’t recall seeing this in any of the lectures up to this point or in the 2017 paper. Where was this covered?