Neural Question Generation from Text: A Preliminary Study

By Qingyu Zhou, Nan Yang, Furu Wei, Chuanqi Tan, Hangbo Bao, M. Zhou. 2017

370 citations recorded by the source corpus.

Automatic question generation aims to generate questions from a text passage where the generated questions can be answered by certain sub-spans of the given passage. Traditional methods mainly use rigid heuristic rules to transform a sentence into related questions. In this work, we propose to apply the neural encoder-decoder model to generate meaningful and diverse questions from natural language sentences. The encoder reads the input text and the answer position, to produce an answer-aware input representation, which is fed to the decoder to generate an answer focused question. We conduct a preliminary study on neural question generation from text with the SQuAD dataset, and the experiment results show that our method can produce fluent and diverse questions.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension (2018, background): “Zhou et al. (2017) improved the diversity of the SQuAD data by generating more questions.”
ColdGANs: Taming Language GANs with Cautious Sampling Strategies (2020, background): “Answer-aware Question Generation (QG) [58] is the task wherein, given a text and a target answer, the goal is to generate a relevant question.”
Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering (2019, methodology): “, 2014) for QG tasks, taking answer sentence as input and outputting the question (Du et al., 2017; Zhou et al., 2017), which proved to be better than rulebased methods.”
Just Ask: Learning to Answer Questions from Millions of Narrated Videos (2020, positive): “Moreover, rule-based generation requires the manual creation of rules by experts which is expensive, and has also been recently outperformed by neural question generation [21, 92, 102] as used in our approach.”
Paragraph-level Neural Question Generation with Maxout Pointer and Gated Self-attention Networks (2018, methodology): “QG results are generated by model s2s-a-ct-mp-gsa . only a sentence as context, e.g. (Du et al., 2017; Song et al., 2017; Zhou et al., 2017); when applied to paragraph-level context (Yuan et al., 2017) we observed large gaps compared to state-of-the-art results achieved by using sentence-level…”
Harvesting Paragraph-level Question-Answer Pairs from Wikipedia (2018, background): “Recent work along these lines (Du et al., 2017; Zhou et al., 2017) (see Section 2) has proposed the use of attention-based recurrent neural models trained on the crowdsourced SQuAD dataset (Ra-jpurkar et al., 2016) for question generation.”
QuestEval: Summarization Asks for Fact-based Evaluation (2021, methodology): “For the QG component, we draw on recent work on neural answer-conditional question generation (Zhou et al., 2017).”
Improving Question Answering Model Robustness with Synthetic Adversarial Data Generation (2021, background): “Dataset statistics (see Appendix C), highlight the high percentage of overlapping answers suggesting that existing answer tagging methods (Zhou et al., 2017; Zhao et al., 2018) might struggle, and models should ideally be capable of handling span overlap.”
G-DAug: Generative Data Augmentation for Commonsense Reasoning (2020, background): “A candidate solution that has shown promise in other tasks, such as reading comprehension, is to augment a human-authored training set with a large set of synthetically-generated examples (Zhou et al., 2017; Du et al., 2017; Zhao et al., 2018a).”
SAFEval: Summarization Asks for Fact-based Evaluation (2021, methodology): “For the QG component, we draw on recent work on neural answer-conditional question generation (Zhou et al., 2017).”
Review-based Question Generation with Adaptive Instance Transfer and Augmentation (2019, methodology): “The preliminary neural QG models (Du et al., 2017; Zhou et al., 2017; Du and Cardie, 2017) outperform the rule-based methods relying on hand-craft features, and thereafter various models have been proposed to further improve the performance via incorporating question type (Dong et al., 2018),…”
Learning with Limited Data for Multilingual Reading Comprehension (2019, methodology): “For QG and AE implementations, we follow the setting of NQG model (Zhou et al., 2017).”
A Unified Query-based Generative Model for Question Generation and Question Answering (2017, methodology): “Here, we follow Du, Shao, and Cardie (2017) and Zhou et al. (2017) to conduct experiments on the accessible part as our entire dataset.”
GLGE: A New General Language Generation Evaluation Benchmark (2020, background): “Compared with answer-agnostic question generation tasks that can generate lots of reasonable questions, answer-aware question generation (Zhou et al., 2017) is asked to generate a question asks towards the given answer span based on a given text passage or document.”
Identifying Where to Focus in Reading Comprehension for Neural Question Generation (2017, background): “In particular, prior work focuses almost exclusively on sentence-level question generation: given a text passage, assume that all sentences contain a question-worthy concept and generate one or more questions for each (Heilman and Smith, 2010; Du et al., 2017; Zhou et al., 2017).”
Question Answering and Question Generation as Dual Tasks (2017, background): “As the answer is a text span from the sentence/passage, it is helpful to incorporate the position of the answer span (Zhou et al. 2017).”
Two Birds, One Stone: A Simple, Unified Model for Text Generation from Structured and Unstructured Data (2019, methodology): “Thus, Du et al. (2017) and Zhou et al. (2017) re-divide available data into training, validation, and test sets, which we call split-1 and split-2, respectively.”
Teaching Machines to Ask Questions (2018, methodology): “Learning the variety of a valid question is an important but overlooked problem in many existing methods [Zhou et al. , 2017; Yuan et al. , 2017].”
Answer-focused and Position-aware Neural Question Generation (2018, methodology): “Pointer-generator Model) Our baseline model is an attention-based pointer-generator model (See et al., 2017) enhanced with various rich features proposed by (Zhou et al., 2017).”