Abstract
Existing algorithms for the Generation of Referring Expressions (GRE) aim at generating descriptions that allow a hearer to identify its intended referent uniquely; the length of the expression is also considered, usually as a secondary issue. We explore the possibility of making the trade-off between these two factors more explicit, via a general cost function which scores these two aspects separately. We sketch some more complex phenomena which might be amenable to this treatment.
| Original language | English |
|---|---|
| Title of host publication | COLING/ACL 2006 |
| Subtitle of host publication | INLG-06 - 4th International Natural Language Generation Conference, Proceedings of the Conference |
| Pages | 89-91 |
| Number of pages | 3 |
| Publication status | Published - 2011 |
| Externally published | Yes |
| Event | 4th International Natural Language Generation Conference, INLG-06 - Sydney, NSW, Australia Duration: 15 Jul 2006 → 16 Jul 2006 |
Conference
| Conference | 4th International Natural Language Generation Conference, INLG-06 |
|---|---|
| Country/Territory | Australia |
| City | Sydney, NSW |
| Period | 15/07/06 → 16/07/06 |