Who Wrote This Line? Evaluating the Detection of LLM-Generated Classical Chinese Poetry

TL;DR

This paper introduces ChangAn, a large benchmark dataset for detecting AI-generated classical Chinese poetry, revealing current detectors' limitations and emphasizing the need for specialized evaluation tools.

Contribution

The paper presents ChangAn, a comprehensive benchmark dataset for LLM-generated classical Chinese poetry detection, and systematically evaluates existing detectors' performance on this challenging task.

Findings

Current detectors are unreliable for classical Chinese poetry detection.

The dataset contains over 30,000 poems, including human and AI-generated works.

Evaluation shows significant performance gaps in existing detection methods.

Abstract

The rapid development of large language models (LLMs) has extended text generation tasks into the literary domain. However, AI-generated literary creations has raised increasingly prominent issues of creative authenticity and ethics in literary world, making the detection of LLM-generated literary texts essential and urgent. While previous works have made significant progress in detecting AI-generated text, it has yet to address classical Chinese poetry. Due to the unique linguistic features of classical Chinese poetry, such as strict metrical regularity, a shared system of poetic imagery, and flexible syntax, distinguishing whether a poem is authored by AI presents a substantial challenge. To address these issues, we introduce ChangAn, a benchmark for detecting LLM-generated classical Chinese poetry that containing total 30,664 poems, 10,276 are human-written poems and 20,388 poems are generated by four popular LLMs. Based on ChangAn, we conducted a systematic evaluation of 12 AI detectors, investigating their performance variations across different text granularities and generation strategies. Our findings highlight the limitations of current Chinese text detectors, which fail to serve as reliable tools for detecting LLM-generated classical Chinese poetry. These results validate the effectiveness and necessity of our proposed ChangAn benchmark. Our dataset and code are available at https://github.com/VelikayaScarlet/ChangAn.