Paper: Generative Alignment and Semantic Parsing for Learning from Ambiguous Supervision

ACL ID C10-2062
Title Generative Alignment and Semantic Parsing for Learning from Ambiguous Supervision
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010
Authors

We present a probabilistic generative model for learning semantic parsers from ambiguous supervision. Our approach learns from natural language sentences paired with world states consisting of multiple potential logical meaning repre- sentations. It disambiguates the mean- ing of each sentence while simultane- ously learning a semantic parser that maps sentences into logical form. Compared to a previous generative model for se- mantic alignment, it also supports full semantic parsing. Experimental results on the Robocup sportscasting corpora in both English and Korean indicate that our approach produces more accurate se- mantic alignments than existing methods and also produces competitive semantic parsers and improved language genera- tors.