Lingxi: A Diversity-aware Chinese Modern Poetry Generation System

TL;DR

Lingxi is a Chinese poetry generation system that enhances creativity and diversity by introducing novel sampling algorithms to produce more original and contextually relevant poetry.

Contribution

It introduces NS-RH, a novel sampling method that increases diversity and novelty in poetry generation, and a semantic-similarity-based rejection sampling to improve contextual relevance.

Findings

NS-RH significantly increases the novelty of generated poetry.

Randomizing the high-frequency words maintains fluency while boosting creativity.

Semantic-similarity-based rejection improves relevance to input titles.

Abstract

Poetry generation has been a difficult task in natural language processing. Unlike plain neural text generation tasks, poetry has a high requirement for novelty, since an easily-understood sentence with too many high frequency words might not be considered as poetic, while adequately ambiguous sentences with low frequency words can possibly be novel and creative. Inspired by this, we present Lingxi, a diversity-aware Chinese modern poetry generation system. We propose nucleus sampling with randomized head (NS-RH) algorithm, which randomizes the high frequency part ("head") of the predicted distribution, in order to emphasize on the "comparatively low frequency" words. The proposed algorithm can significantly increase the novelty of generated poetry compared with traditional sampling methods. The permutation of distribution is controllable by tuning the filtering parameter that determines the "head" to permutate, achieving diversity-aware sampling. We find that even when a large portion of filtered vocabulary is randomized, it can actually generate fluent poetry but with notably higher novelty. We also propose a semantic-similarity-based rejection sampling algorithm, which creates longer and more informative context on the basis of the short input poetry title while maintaining high semantic similarity to the title, alleviating the off-topic issue.