Semantic Graphs for Generating Deep Questions
arXiv:2004.12704
Abstract
This paper proposes the problem of Deep Question Generation (DQG), which aims to generate complex questions that require reasoning over multiple pieces of information of the input passage. In order to capture the global structure of the document and facilitate reasoning, we propose a novel framework which first constructs a semantic-level graph for the input document and then encodes the semantic graph by introducing an attention-based GGNN (Att-GGNN). Afterwards, we fuse the document-level and graph-level representations to perform joint training of content selection and question decoding. On the HotpotQA deep-question centric dataset, our model greatly improves performance over questions requiring reasoning over multiple facts, leading to state-of-the-art performance. The code is publicly available at https://github.com/WING-NUS/SG-Deep-Question-Generation.
ACL 2020 (long paper)
References in corpus (12)
- Neural Machine Translation by Jointly Learning to Align and Translate
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- Incorporating Copying Mechanism in Sequence-to-Sequence Learning
- Simple BERT Models for Relation Extraction and Semantic Role Labeling
- Modeling Coverage for Neural Machine Translation
- HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
- Learning to Ask: Neural Question Generation for Reading Comprehension
- Recent Advances in Neural Question Generation
- Multi-Hop Knowledge Graph Reasoning with Reward Shaping
- Neural Question Generation from Text: A Preliminary Study
- Harvesting Paragraph-Level Question-Answer Pairs from Wikipedia
- Self-Assembling Modular Networks for Interpretable Multi-Hop Reasoning