Neural machine translation app: http://104.131.78.120/pic.twitter.com/8nWAMcowLw
-
Only tried it on very short, basic sentences with some simple semantic ambiguity that would be obviously clear to humans. 0% success so far.
Maybe it's a training data issue. Is this approach supposed to be able to deal with disambiguation?
-
-
that's a small model we trained in 2014. kinda out-of-date
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
in principle - no. Because sentences do not convey full context. Google has a lot of data. No success, though some progress.
-
even trivial n-gram models can do simple disambiguation. It's not a hard problem, at least not at that level.
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.