Generating Major Types of Chinese Classical Poetry in a Uniformed Framework

TL;DR

This paper introduces a GPT-2 based framework for generating various types of Chinese classical poetry, emphasizing form control and high-quality content, validated through experiments and integration into an existing poetry system.

Contribution

It proposes a unified format and form-stressed weighting method in GPT-2 for controlled Chinese poetry generation, especially for longer forms.

Findings

High-quality generation of major Chinese poetic forms

Effective form control in generated poems

Successful integration into Jiuge system

Abstract

Poetry generation is an interesting research topic in the field of text generation. As one of the most valuable literary and cultural heritages of China, Chinese classical poetry is very familiar and loved by Chinese people from generation to generation. It has many particular characteristics in its language structure, ranging from form, sound to meaning, thus is regarded as an ideal testing task for text generation. In this paper, we propose a GPT-2 based uniformed framework for generating major types of Chinese classical poems. We define a unified format for formulating all types of training samples by integrating detailed form information, then present a simple form-stressed weighting method in GPT-2 to strengthen the control to the form of the generated poems, with special emphasis on those forms with longer body length. Preliminary experimental results show this enhanced model can generate Chinese classical poems of major types with high quality in both form and content, validating the effectiveness of the proposed strategy. The model has been incorporated into Jiuge, the most influential Chinese classical poetry generation system developed by Tsinghua University (Guo et al., 2019).