Supplementary Material for
Semantic Parsing of Ambiguous Input through Paraphrasing and Verification
Philip Arthur, Graham Neubig, Sakriani Sakti, Tomoki Toda, Satoshi Nakamura
Nara Institute of Science and Technology - Augmented Human Communication Lab


We propose a new method for semantic parsing of ambiguous and ungrammatical input,such as search queries. We do so by building on an existing semantic parsing framework that uses synchronous context free grammars (SCFG) to jointly model the input sentence and output meaning representation. We generalize this SCFG framework to allow not one, but multiple outputs. Using this formalism, we construct a grammar that takes an ambiguous input string and jointly maps it into both a meaning representation and a natural language paraphrase that is less ambiguous than the original input. This paraphrase can be used to disambiguate the meaning representation via verification using a language mode that calculates the probability of each paraphrase.


To reproduce all the experiments you will need to first install:

  • Travatar: Syntax based machine translation decoder for tree-to-string and hierarchical translation.
  • Pialign: Phrasal aligner tools.
  • Stanford Parser: To perform all preprocessings for building a language model
  • SWI-Prolog: Prolog interpreter to execute Geoquery database.
  • Lamtram: A tool to build neural-network language model.
  • GIZA++: Word aligner based on IBM models.
  • Letrac: My own implementation of Wong&Mooney, 2007 and Li et al., 2013.

After installing all these tools, make sure you link them inside config.ini later.


  • Code+Dataset Scripts and dataset can be downloaded for this paper released on Github.
  • Keyword Data In case you are quickly looking for our keyword data.

If you have any problems using the scripts, please feel free to contact philip.arthur.om0 (!