Hi,
the short answer to your problem would be, that the typical encoder
decoder models are not really meant to do what you want it to do, there
is however interesting new work on archive:
https://arxiv.org/abs/1611.01874
which could exactly solve your problem. However, I am always weary of
results of that particular group of researchers. It seems reproducing
their results for anything but Chinese does not really work, also their
train sets are really small, so it is not clear what the effects really
are. Maybe those models are just dealing better with smaller data.
Best,
Marcin
W dniu 24/11/16 o 10:22, Nat Gillin pisze:
Dear Moses Community,
This seems to be prickly topic to discuss but my experiments on a
different kind of data set than WMT or WAT (workshop for asian
translation) has not been able to achieve the stella scores that the
recent advancement in MT has been reporting.
Using state-of-art encoder-attention-decoder framework, just by
running things like lamtram or tensorflow, I'm unable to beat Moses'
scores from sentences that appears both in the train and test data.
Imagine it's a translator using MT and somehow he/she has translated
the sentence before and just wants the exact translation. A TM would
solve the problem and Moses surely could emulate the TM but NMT tends
to go overly creative and produces something else. Although it is
consistent in giving the same output for the same sentence, it's just
unable to regurgitate the sentence that was seen in the training data.
In that matter, Moses does it pretty well.
For sentences that is not in train but in test, NMT does fairly the
same or sometimes better than Moses.
So the question is 'has anyone encounter similar problems?' Is the
solution simply to do a fetch in the train set before translating? Or
a system/output chooser to rerank outputs?
Are there any other ways to resolve such a problem? What could have
happened such that NMT is not "remembering"? (Maybe it needs some
memberberries)
Any tips/hints/discussion on this is much appreciated.
Regards,
Nat
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
