<div dir="ltr">Thanks, Stephan.<div>Emily</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, Apr 29, 2014 at 1:59 PM, Stephan Oepen <span dir="ltr"><<a href="mailto:oe@ifi.uio.no" target="_blank">oe@ifi.uio.no</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">hi again (cleaning up my inbox),<br>
<div class=""><br>
> In fact, it seems that while adding :task :transfer blocks the rules from<br>
> being used as post-transfer fix-up rules, it also blocks them (in this<br>
> set-up) from being used as transfer rules.<br>
<br>
</div>indeed; in the interactive MT setup, the ‘rephrase’ function will be<br>
looking to invoke a transfer grammar flagged for the task :paraphrase<br>
(or not flagged at all, as you used to do). arguably, the<br>
pre-generation fix-up phase should be more selective about which rules<br>
to use, i.e. refuse to match on transfer grammars whose :task is<br>
underspecified (as you used to do).<br>
<br>
best, oe<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>Emily M. Bender<br>Associate Professor<br>Department of Linguistics<br>Check out CLMS on facebook! <a href="http://www.facebook.com/uwclma" target="_blank">http://www.facebook.com/uwclma</a><br>
</div>