Correcting Semantic Parses with Natural Language through Dynamic Schema Encoding

Parker Glenn, Parag Pravin Dakle, Preethi Raghavan
5/31/2023
cs.CL

No official code found. Best implementation: parkervg/DestT5 (3 stars, last commit 2024-02-15).

Abstract

In addressing the task of converting natural language to SQL queries, there are several semantic and syntactic challenges. It becomes increasingly important to understand and remedy the points of failure as the performance of semantic parsing systems improve. We explore semantic parse correction with natural language feedback, proposing a new solution built on the success of autoregressive decoders in text-to-SQL tasks. By separating the semantic and syntactic difficulties of the task, we show that the accuracy of text-to-SQL parsers can be boosted by up to 26% with only one turn of correction with natural language. Additionally, we show that a T5-base model is capable of correcting the errors of a T5-large model in a zero-shot, cross-parser setting.

View on arXivView PDF

Code Implementations(1)

parkervg/DestT5Community100%

Code & Dataset for DestT5 (NLP for ConvAI, ACL 2023)

30PythonMay 30, 20232 years ago

Similar Papers

Learning to Simulate Natural Language Feedback for Interactive Semantic Parsing

86% match1 repo52023

On Robustness of Prompt-based Semantic Parsing with Large Pre-trained Language Model: An Empirical Study on Codex

83% match2023

Metasql: A Generate-then-Rank Framework for Natural Language to SQL Translation

83% match1 repo272024

The Role of Output Vocabulary in T2T LMs for SPARQL Semantic Parsing

83% match1 repo42023

TrustSQL: Benchmarking Text-to-SQL Reliability with Penalty-Based Scoring

83% match2024

R³-NL2GQL: A Model Coordination and Knowledge Graph Alignment Approach for NL2GQL

82% match1 repo1032023

Cite this paper

@article{glenn2023correcting,
  title  = {Correcting Semantic Parses with Natural Language through Dynamic Schema Encoding},
  author = {Parker Glenn and Parag Pravin Dakle and Preethi Raghavan},
  year   = {2023},
  eprint = {2305.19974},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.19974v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.19974.svg)](https://codeofpaper.com/paper/2305.19974)

Discussion