[developers] More ACE/LKB divergences (underspecified generation)

Francis Bond bond at ieee.org
Sun Mar 8 05:22:16 CET 2015


--disable-subsumption-test allowed us to (over) generate usefully, thanks.
We added it to AceOptions.



On Fri, Mar 6, 2015 at 9:54 AM, Emily M. Bender <ebender at uw.edu> wrote:

> Thanks, Woodley.  I'll definitely try that when we get to ace. For now,
> it's all LKB,
> though, and it seems like that backing off isn't happening.
>
> Emily
>
> On Thu, Mar 5, 2015 at 5:48 PM, Woodley Packard <sweaglesw at sweaglesw.org>
> wrote:
>
>> On Mar 5, 2015, at 10:18 AM, "Emily M. Bender" <ebender at uw.edu> wrote:
>> > Currently, with the default aspect inserted,
>> > we're getting no generator outputs for adjective examples.  It seems
>> like the
>> > more forgiving setting described for the LKB ought to allow that---is
>> there
>> > something I need to do to invoke this setting?
>>
>>
>> I'm not sure if you are asking about ACE or LKB here, but ACE has an
>> option --disable-subsumption-test that does what it says.  This is not
>> quite the same as the LKB mode applying it by default but backing off if no
>> results would have been produced.
>>
>> -Woodley
>
>
>
>
> --
> Emily M. Bender
> Professor, Department of Linguistics
> Check out CLMS on facebook! http://www.facebook.com/uwclma
>



-- 
Francis Bond <http://www3.ntu.edu.sg/home/fcbond/>
Division of Linguistics and Multilingual Studies
Nanyang Technological University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.delph-in.net/archives/developers/attachments/20150308/58fd783a/attachment.html>


More information about the developers mailing list