Can ChatGPT Really Understand Modern Chinese Poetry?

TL;DR

This paper evaluates ChatGPT's ability to understand modern Chinese poetry through a comprehensive framework, revealing strengths in interpretation alignment but weaknesses in capturing poeticity, and sets a foundation for future research.

Contribution

It introduces a novel evaluation framework for assessing ChatGPT's understanding of modern poetry, involving expert collaboration and multi-dimensional analysis.

Findings

ChatGPT's interpretations align with original poets' intents in over 73% of cases.

Its understanding of poeticity is less satisfactory.

The framework effectively assesses LLMs' poetry comprehension.

Abstract

ChatGPT has demonstrated remarkable capabilities on both poetry generation and translation, yet its ability to truly understand poetry remains unexplored. Previous poetry-related work merely analyzed experimental outcomes without addressing fundamental issues of comprehension. This paper introduces a comprehensive framework for evaluating ChatGPT's understanding of modern poetry. We collaborated with professional poets to evaluate ChatGPT's interpretation of modern Chinese poems by different poets along multiple dimensions. Evaluation results show that ChatGPT's interpretations align with the original poets' intents in over 73% of the cases. However, its understanding in certain dimensions, particularly in capturing poeticity, proved to be less satisfactory. These findings highlight the effectiveness and necessity of our proposed framework. This study not only evaluates ChatGPT's ability to understand modern poetry but also establishes a solid foundation for future research on LLMs and their application to poetry-related tasks.