Few-shot Question Generation for Reading Comprehension
Yin Poon, John Sie Yuen Lee, Yuylam hkmu.edu.hk Yuylam hkmu.edu.hk, Wlsuen hkmu.edu.hk Wlsuen hkmu.edu.hk, Eong hkmu.edu.hk Eong hkmu.edu.hk, Skwchu hkmu.edu.hk Skwchu hkmu.edu.hk · 2024
According to the internationally recognized PIRLS (Progress in International Reading Literacy Study) assessment standards, reading comprehension questions should require not only information retrieval, but also higher-order processes such as inferencing, interpreting and evaluation.However, these kinds of questions are often not available in large quantities for training question generation models.This paper investigates whether pre-trained Large Language Models (LLMs) can produce higherorder questions.Human assessment on a Chinese dataset shows that few-shot LLM prompting generates more usable and higher-order questions than two competitive neural baselines.