Logical Parsing from Natural Language Based on a Neural Translation Model
This work addresses the problem of building general semantic parsers without reliance on hand-crafted features, which is significant for natural language interface systems, though it is incremental as it builds on existing neural methods.
The paper tackles semantic parsing for natural language interfaces by proposing a neural translation model that learns from denotations, using a Seq2Seq approach with attention and dynamic programming to infer logical forms. It demonstrates success in an arithmetic domain, learning word meanings, compositionality, and operation orders simultaneously.
Semantic parsing has emerged as a significant and powerful paradigm for natural language interface and question answering systems. Traditional methods of building a semantic parser rely on high-quality lexicons, hand-crafted grammars and linguistic features which are limited by applied domain or representation. In this paper, we propose a general approach to learn from denotations based on Seq2Seq model augmented with attention mechanism. We encode input sequence into vectors and use dynamic programming to infer candidate logical forms. We utilize the fact that similar utterances should have similar logical forms to help reduce the searching space. Under our learning policy, the Seq2Seq model can learn mappings gradually with noises. Curriculum learning is adopted to make the learning smoother. We test our method on the arithmetic domain which shows our model can successfully infer the correct logical forms and learn the word meanings, compositionality and operation orders simultaneously.