TegTok: Augmenting Text Generation via Task-specific and Open-world Knowledge

TL;DR

TegTok enhances text generation by integrating task-specific and open-world knowledge through a unified retrieval and injection framework, improving performance on dialogue and question generation tasks.

Contribution

The paper introduces a novel framework that combines task-specific and open-world knowledge for improved text generation using knowledge retrieval and injection.

Findings

Outperforms baseline models on dialogue and question generation tasks.

Effectively integrates two types of knowledge into PLMs for better generation.

Demonstrates improved quality and informativeness of generated texts.

Abstract

Generating natural and informative texts has been a long-standing problem in NLP. Much effort has been dedicated into incorporating pre-trained language models (PLMs) with various open-world knowledge, such as knowledge graphs or wiki pages. However, their ability to access and manipulate the task-specific knowledge is still limited on downstream tasks, as this type of knowledge is usually not well covered in PLMs and is hard to acquire. To address the problem, we propose augmenting TExt Generation via Task-specific and Open-world Knowledge (TegTok) in a unified framework. Our model selects knowledge entries from two types of knowledge sources through dense retrieval and then injects them into the input encoding and output decoding stages respectively on the basis of PLMs. With the help of these two types of knowledge, our model can learn what and how to generate. Experiments on two text generation tasks of dialogue generation and question generation, and on two datasets show that our method achieves better performance than various baseline models.