Reinforced Dynamic Reasoning for Conversational Question Generation

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 2114–2124 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 2114 Reinforced Dynamic Reasoning for Conversational Question Generation Boyuan Pan1∗, Hao Li1, Ziyu Yao2, Deng Cai1,3, Huan Sun2 1State Key Lab of CAD&CG, Zhejiang University 2The Ohio State University 3Alibaba-Zhejiang University Joint Institute of Frontier Technologies {panby, haolics, dcai}@zju.edu.cn {yao.470, sun.397}@osu.edu Abstract This paper investigates a new task named Conversational Question Generation (CQG) which is to generate a question based on a pas- sage and a conversation history (i.e., previous turns of question-answer pairs). CQG is a cru- cial task for developing intelligent agents that can drive question-answering style conversa- tions or test user understanding of a given pas- sage. Towards that end, we propose a new ap- proach named Reinforced Dynamic Reason- ing (ReDR) network, which is based on the general encoder-decoder framework but incor- porates a reasoning procedure in a dynamic manner to better understand what has been asked and what to ask next about the pas- sage. To encourage producing meaningful questions, we leverage a popular question an- swering (QA) model to provide feedback and fine-tune the question generator using a rein- forcement learning mechanism. Empirical re- sults on the recently released CoQA dataset demonstrate the effectiveness of our method in comparison with various baselines and model variants. Moreover, to show the applicability of our method, we also apply it to create multi- turn question-answering conversations for pas- sages in SQuAD. 1 Introduction In this work, we study a novel task of conversa- tional question generation (CQG) which is given a passage and a conversation history (i.e., previ- ous turns of question-answer pairs), to generate the next question. CQG is an important task in its own right for measuring the ability of machines to lead a question-answering style conversation. It can serve as an essential component of intelligent so- cial bots or tutoring systems, asking meaningful ∗Work done while visiting the Ohio State University. Shelly is in second grade. She is a new student at her school. Shelly's family has lived in many different places. Shelly was born in Florida. Her family moved to Tennessee when she was two years old. When she was four years old, they moved to Texas. They moved from there to Arizona, where they now live. Q1: What grade is Shelly in ? A1: second R1: Shelly is in second grade. Q2: Was she a new student ? A2: Yes R2: She is a new student at her school. Q3: Where did she move at 2 years old ? A2: Tennessee R3: Her family moved to Tennessee when she was two years old. Figure 1: An example from the CoQA dataset. Each turn contains a question (Q) and an answer (A). The dataset also provides a rationale (R) (i.e., a text span from the passage) to support each answer. and coherent questions to engage users or test stu- dent understanding about a certain topic. On …

Content truncated to 500 words