This repository provides a curated list of essential research papers for Neural Question Generation (NQG). It serves as a comprehensive resource for researchers and practitioners looking to understand the state-of-the-art in generating questions from text, knowledge graphs, and visual inputs. The list is organized by topic, covering models, applications, evaluation, and resources.

How It Works

The repository itself is a static list of papers, not a runnable codebase. It categorizes seminal and recent works in NQG, providing links to their respective publications. The organization follows a logical progression from foundational models to specialized applications and evaluation metrics, offering a structured learning path for newcomers to the field.

Highlighted Details

• Extensive coverage of NQG techniques, including basic Seq2Seq, answer encoding, linguistic features, reinforcement learning, content selection, and pre-trained models.
• Detailed sections on diverse applications such as difficulty-controllable QG, conversational QG, visual QG, distractor generation, and cross-lingual QG.
• Includes papers on evaluation metrics and relevant datasets/toolkits for NQG research.
• Papers are linked to their original sources, often including arXiv, conference proceedings, and sometimes code repositories.

Maintenance & Community

This is a community-contributed list, with contributions noted from Liangming Pan, Yuxi Xie, and Yunxiang Zhang. The list appears to be updated periodically to include recent advancements.

Licensing & Compatibility

The repository itself does not have a specified license, but it links to academic papers, which are subject to their own publication licenses. Usage of the paper content is governed by the original publishers.

Limitations & Caveats

This repository is a bibliography and does not contain any code or executable models. It is a curated list of papers, not a framework or library for implementing NQG systems. Users will need to find and implement the models described in the papers themselves.