# Recent Advances in Neural Question Generation

**Authors:** Liangming Pan, Wenqiang Lei, Tat-Seng Chua, Min-Yen Kan  
**arXiv:** [1905.08949](https://arxiv.org/abs/1905.08949)  
**Date:** 2019-05-22  
**Summary page:** https://www.summarizepaper.com/en/paper/1905.08949v1/

## Key points

- Neural Question Generation (NQG) has been a subject of much research and development in recent years.
- Integration of a wider variety of inputs has led to the generation of questions that require higher levels of cognition.
- NQG is a bellwether for Natural Language Processing (NLP), highlighting how human intelligence embodies skills such as curiosity and integration.
- Liangming Pan, Wenqiang Lei, Tat-Seng Chua, and Min-Yen Kan conducted a survey examining corpora, methodologies, and evaluation methods to provide a comprehensive overview of NQG.
- Emerging trends in NQG's learning paradigms, input modalities, and cognitive levels considered by researchers are revealed in the survey.
- Promising directions for future research include incorporating external knowledge sources into question generation models or developing more sophisticated evaluation metrics to assess the quality of generated questions.
- This survey emphasizes the importance of NQG as an area of study within NLP and provides valuable insights into its current state and potential future developments.

## Summary

In recent years, Neural Question Generation (NQG) has been the subject of much research and development. One notable trend in this field is the integration of a wider variety of inputs, which has led to the generation of questions that require higher levels of cognition. This development has positioned NQG as a bellwether for Natural Language Processing (NLP), highlighting how human intelligence embodies skills such as curiosity and integration. To provide a comprehensive overview of NQG, Liangming Pan, Wenqiang Lei, Tat-Seng Chua, and Min-Yen Kan have conducted a survey examining corpora, methodologies, and evaluation methods. Their findings reveal emerging trends in NQG's learning paradigms, input modalities, and cognitive levels considered by researchers. The authors note that there are several promising directions for future research in this area. For instance, they suggest exploring how to incorporate external knowledge sources into question generation models or developing more sophisticated evaluation metrics to assess the quality of generated questions. Overall, this survey emphasizes the importance of NQG as an area of study within NLP and provides valuable insights into its current state and potential future developments.