Question Answering and Question Generation as Dual Tasks

By Duyu Tang, Nan Duan, Tao Qin, M. Zhou. 2017

190 citations recorded by the source corpus.

We study the problem of joint question answering (QA) and question generation (QG) in this paper. Our intuition is that QA and QG have intrinsic connections and these two tasks could improve each other. On one side, the QA model judges whether the generated question of a QG model is relevant to the answer. On the other side, the QG model provides the probability of generating a question given the answer, which is a useful evidence that in turn facilitates QA. In this paper we regard QA and QG as dual tasks. We propose a training framework that trains the models of QA and QG simultaneously, and explicitly leverages their probabilistic correlation to guide the training process of both models. We implement a QG model based on sequence-to-sequence learning, and a QA model based on recurrent neural network. As all the components of the QA and QG models are differentiable, all the parameters involved in these two models could be conventionally learned with back propagation. We conduct experiments on three datasets. Empirical results show that our training framework improves both QA and QG tasks. The improved QA model performs comparably with strong baseline approaches on all three datasets.

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

Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering (2019, background): “Some other works enabled semi-supervised QA with the help of QG. Tang et al. (2017) applied the “dual learning” al-gorithm (He et al., 2016) to learn QA and QG jointly with unlabeled texts.”
Towards a Better Metric for Evaluating Question Generation Systems (2018, background): “A few models (Wang et al., 2017; Tang et al., 2017) treat question answering (QA) and question generation (QG) as complementary tasks and focus on jointly training for these two tasks.”
Achieving Human Parity on Automatic Chinese to English News Translation (2018, background): “Ever since dual learning was proposed, it has been successfully applied to various real-world problems such as question answering [35], image classification [47], image segmentation [25], image to image translation [50, 52, 24], face attribute manipulation [31], and machine translation [19, 43, 23, ”
Paragraph-level Neural Question Generation with Maxout Pointer and Gated Self-attention Networks (2018, background): “(Tang et al., 2017; Wang et al., 2017a) proposed joint models to address QG and question answering together.”
Generating Fact Checking Briefs (2020, neutral): “Unlike previous question generation models (Du et al., 2017; Duan et al., 2017; Tang et al., 2017; Zhao et al., 2018) that generate based on an answer, we treat question generation closer to structured planning — laying out the format for the entire brief.”
Visual Question Generation as Dual Task of Visual Question Answering (2017, background): “introduce the idea to QA area, where question generation is modeled as dual task of QA, and leverage the probabilistic correlation between QA and QG to guide the training [31].”
Learning with Limited Data for Multilingual Reading Comprehension (2019, methodology): “Combining models of QG and QA: Several works (Wang et al., 2017; Tang et al., 2017; Tang et al.) propose a framework jointly optimizing QG and QA model as dual.”
Answer-focused and Position-aware Neural Question Generation (2018, background): “Tang et al. (2017) model question answering and question generation as dual tasks.”
Learning to Collaborate for Question Answering and Asking (2018, methodology): “Therefore, we follow (Tang et al., 2017) and split the original validation set into the dev set and the test set.”
Simple and Effective Semi-Supervised Question Answering (2018, background): “Several approaches have been suggested for generating nat-ural questions (Tang et al., 2017; Subramanian et al., 2017; Song et al., 2017), however none of them show a signiﬁcant improvement of using the generated questions in a semi-supervised setting.”
Two Birds, One Stone: A Simple, Unified Model for Text Generation from Structured and Unstructured Data (2019, background): “…yet challenging task in NLP that has received growing attention due to its application in improving question answering systems (Duan et al., 2017; Tang et al., 2017, 2018), providing material for educational purposes (Heilman and Smith, 2010), and helping conversational systems to start and…”
Code Generation as a Dual Task of Code Summarization (2019, background): “Tang et al. [2017] treated question answering and question generation as a dual task and proved the effectiveness of dual supervised learning.”
A Recurrent BERT-based Model for Question Generation (2019, background): “Further, (Tang et al., 2017; Wang et al., 2017) proposed joint models to address QG and question answering as a multi-task learning setting.”
Neural-Symbolic Solver for Math Word Problems with Auxiliary Tasks (2021, background): “, 2017), question answering (Tang et al., 2017), visual question answering (Li et al.”
Learning Dual Retrieval Module for Semi-supervised Relation Extraction (2019, background): “For example, English-French translation with French-English translation [14], image classification with image generation [35], question answering and question generation [32] can be treated together to leverage the duality of two tasks.”
A Unified Query-based Generative Model for Question Generation and Question Answering (2017, methodology): “For question generation (QG), our work extends previous work (Du, Shao, and Cardie, 2017; Zhou et al., 2017; Yang et al., 2017; Subramanian et al., 2017; Tang et al., 2017; Wang, Yuan, and Trischler, 2017; Yuan et al., 2017) by performing query understanding.”
Learning to Generate Questions by LearningWhat not to Generate (2019, methodology): “One of the key applications of question generation is to automatically produce question-answer pairs to enhance machine reading comprehension systems [8, 34, 35, 38].”
CoaCor: Code Annotation for Code Retrieval with Reinforcement Learning (2019, background): “Various kinds of machine-machine collaboration mechanisms have been studied in many scenarios [12, 15, 28, 55, 59].”