Abstract
We propose a new algorithm to approximately extract top-scoring hypotheses from a hypergraph when the score includes an N-gram language model. In the popular cube pruning algorithm, every hypothesis is annotated with boundary words and permitted to recombine only if all boundary words are equal. However, many hypotheses share some, but not all, boundary words. We use these common boundary words to group hypotheses and do so recursively, resulting in a tree of hypotheses. This tree forms the basis for our new search algorithm that iteratively refines groups of boundary words on demand. Machine translation experiments show our algorithm makes translation 1.50 to 3.51 times as fast as with cube pruning in common cases.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies |
| Subtitle of host publication | Human Language Technologies, Proceedings of the Main Conference |
| Publisher | Association for Computational Linguistics (ACL) |
| Pages | 958-968 |
| Number of pages | 11 |
| ISBN (Electronic) | 9781937284473 |
| Publication status | Published - 9 Jun 2013 |
| Event | 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013 - Atlanta, United States Duration: 9 Jun 2013 → 14 Jun 2013 |
Conference
| Conference | 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013 |
|---|---|
| Country/Territory | United States |
| City | Atlanta |
| Period | 9/06/13 → 14/06/13 |