Paper: Adjoining Tree-to-String Translation

ACL ID P11-1128
Title Adjoining Tree-to-String Translation
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2011

We introduce synchronous tree adjoining grammars (TAG) into tree-to-string transla- tion, which converts a source tree to a target string. Without reconstructing TAG deriva- tions explicitly, our rule extraction algo- rithm directly learns tree-to-string rules from aligned Treebank-style trees. As tree-to-string translation casts decoding as a tree parsing problem rather than parsing, the decoder still runs fast when adjoining is included. Less than 2 times slower, the adjoining tree-to- string system improves translation quality by +0.7 BLEU over the baseline system only al- lowing for tree substitution on NIST Chinese- English test sets.