May the Force Be with Your Copy Mechanism: Enhanced Supervised-Copy Method for Natural Language Generation

TL;DR

This paper introduces a supervised copy mechanism for neural text generation models that improves copying accuracy and abstractness by guiding the model on which words to copy or generate, enhancing performance on data-to-text and summarization tasks.

Contribution

A novel supervised copy network with a redefined objective function that leverages source and target guidance to improve copying decisions in neural sequence models.

Findings

Enhanced copying quality in experiments

Improved abstractness in generated texts

Better performance on data-to-text and summarization tasks

Abstract

Recent neural sequence-to-sequence models with a copy mechanism have achieved remarkable progress in various text generation tasks. These models addressed out-of-vocabulary problems and facilitated the generation of rare words. However, the identification of the word which needs to be copied is difficult, as observed by prior copy models, which suffer from incorrect generation and lacking abstractness. In this paper, we propose a novel supervised approach of a copy network that helps the model decide which words need to be copied and which need to be generated. Specifically, we re-define the objective function, which leverages source sequences and target vocabularies as guidance for copying. The experimental results on data-to-text generation and abstractive summarization tasks verify that our approach enhances the copying quality and improves the degree of abstractness.