Paper: Aligning Formal Meaning Representations with Surface Strings for Wide-Coverage Text Generation

ACL ID W13-2101
Title Aligning Formal Meaning Representations with Surface Strings for Wide-Coverage Text Generation
Venue European Workshop on Natural Language Generation
Session
Year 2013
Authors

Statistical natural language generation from abstract meaning representations presupposes large corpora consisting of text?meaning pairs. Even though such corpora exist nowadays, or could be con- structed using robust semantic parsing, the simple alignment between text and mean- ing representation is too coarse for de- veloping robust (statistical) NLG systems. By reformatting semantic representations as graphs, fine-grained alignment can be obtained. Given a precise alignment at the word level, the complete surface form of a meaning representations can be deduced using a simple declarative rule.