A Globally Normalized Neural Model for Semantic Parsing

TL;DR

This paper introduces a globally normalized neural model for semantic parsing that predicts real-valued scores at each step, overcoming label bias and improving performance on small datasets.

Contribution

It presents a novel globally normalized approach for CFG-based semantic parsing that addresses label bias and enhances accuracy on limited data.

Findings

Outperforms local models on small datasets

No improvement observed on large datasets

Addresses label bias in semantic parsing

Abstract

In this paper, we propose a globally normalized model for context-free grammar (CFG)-based semantic parsing. Instead of predicting a probability, our model predicts a real-valued score at each step and does not suffer from the label bias problem. Experiments show that our approach outperforms locally normalized models on small datasets, but it does not yield improvement on a large dataset.