Informative Text Generation from Knowledge Triples

TL;DR

This paper introduces a memory-augmented model for KB-to-text generation that produces more informative sentences by leveraging learned knowledge beyond input triples, extending the traditional aligned setting.

Contribution

It proposes a novel memory network approach to generate more informative text from knowledge triples, addressing the limitation of existing aligned KB-to-text models.

Findings

The memory-augmented model outperforms baseline methods in generating more informative text.

The approach effectively utilizes learned knowledge to enhance text informativeness.

Experimental results demonstrate the model's ability to generate richer, more comprehensive descriptions.

Abstract

As the development of the encoder-decoder architecture, researchers are able to study the text generation tasks with broader types of data. Among them, KB-to-text aims at converting a set of knowledge triples into human readable sentences. In the original setting, the task assumes that the input triples and the text are exactly aligned in the perspective of the embodied knowledge/information. In this paper, we extend this setting and explore how to facilitate the trained model to generate more informative text, namely, containing more information about the triple entities but not conveyed by the input triples. To solve this problem, we propose a novel memory augmented generator that employs a memory network to memorize the useful knowledge learned during the training and utilizes such information together with the input triples to generate text in the operational or testing phase. We derive a dataset from WebNLG for our new setting and conduct extensive experiments to investigate the effectiveness of our model as well as uncover the intrinsic characteristics of the setting.