On Fri, Mar 30, 2012 at 2:08 PM, Thomas Schoenemann <
[email protected]> wrote:
> Hi everyone,
>
> there is quite a discussion about hierarchical and syntax-based
> approaches here at the moment. For lack of time I could not follow it
> closely, so I hope this question contributes to the discussion rather than
> being superfluous: in terms of {tree/string} -> {tree/string}, how would
> you classify Hiero? My understanding of tree -> {string/tree} is that you
> have a source tree given, which is not the case for Hiero. And
> string->string does in my understanding not involve CFG-like rules at all.
> So Hiero would be string->tree. But this solution is entirely based on
> eliminating the alternatives. I'd prefer to have an expert come up with a
> constructive explanation.
>
> Thanks much!
> Thomas (currently University of Düsseldorf)
>
> P.S: MERT now has an optional positivity constraint.
>
> _______________________________________________
> Moses-support mailing list
> [email protected]
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
>
Also, if you're reading the literature on different types of
syntactically-informed translation models, be aware that many of the
(mostly older) papers use noisy channel terminology.
In those older papers, the authors switch the direction of the model, and
so for what you might think of as a string->tree system, the authors of
such papers say they have a tree->string system. In other words, they write
output->input instead of input->output.
Lane
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
