NL-EDIT: Correcting semantic parse errors through natural language interaction

TL;DR

NL-EDIT is a model that interprets natural language feedback to correct semantic parse errors, significantly improving text-to-SQL parser accuracy with minimal user interaction.

Contribution

The paper introduces NL-EDIT, a novel model for interpreting natural language corrections to improve semantic parsing accuracy in an interactive setting.

Findings

NL-EDIT boosts parser accuracy by up to 20% with one correction turn.

The model effectively interprets natural language feedback for semantic parse correction.

Analysis reveals current limitations and future directions for interactive semantic parsing.

Abstract

We study semantic parsing in an interactive setting in which users correct errors with natural language feedback. We present NL-EDIT, a model for interpreting natural language feedback in the interaction context to generate a sequence of edits that can be applied to the initial parse to correct its errors. We show that NL-EDIT can boost the accuracy of existing text-to-SQL parsers by up to 20% with only one turn of correction. We analyze the limitations of the model and discuss directions for improvement and evaluation. The code and datasets used in this paper are publicly available at http://aka.ms/NLEdit.