Neural Semantic Parsing with Type Constraints for Semi-Structured Tables

This page contains supplementary material for the paper: Neural Semantic Parsing with Type Constraints for Semi-Structured Tables (EMNLP 2017).

The semantic parser was originally written using PNP, a probabilistic neural programming language that uses scala bindings for DyNet. This implementation is available in the wikitables2 branch of the PNP repository.

We are currently re-implementing the parser using AllenNLP, but that code is not ready yet. You can track progress in the wikitables2 branch of that repository.