# Neural Semantic Parsing over Multiple Knowledge-bases

## TL;DR

This paper introduces a multi-knowledge-base neural semantic parser that leverages structural language regularities across domains, significantly improving accuracy and reducing model complexity.

## Contribution

It presents a novel approach to train a single sequence-to-sequence parser over multiple KBs with domain encoding, achieving state-of-the-art results.

## Key findings

- Improved parsing accuracy from 75.6% to 79.6%.
- Achieved 7x reduction in model parameters.
- State-of-the-art performance on the Overnight dataset.

## Abstract

A fundamental challenge in developing semantic parsers is the paucity of strong supervision in the form of language utterances annotated with logical form. In this paper, we propose to exploit structural regularities in language in different domains, and train semantic parsers over multiple knowledge-bases (KBs), while sharing information across datasets. We find that we can substantially improve parsing accuracy by training a single sequence-to-sequence model over multiple KBs, when providing an encoding of the domain at decoding time. Our model achieves state-of-the-art performance on the Overnight dataset (containing eight domains), improves performance over a single KB baseline from 75.6% to 79.6%, while obtaining a 7x reduction in the number of model parameters.