Learning to Paraphrase for Question Answering

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 875–886 Copenhagen, Denmark, September 7–11, 2017. c⃝2017 Association for Computational Linguistics Learning to Paraphrase for Question Answering Li Dong† and Jonathan Mallinson† and Siva Reddy‡ and Mirella Lapata† † ILCC, School of Informatics, University of Edinburgh ‡ Computer Science Department, Stanford University li.dong@ed.ac.uk, J.Mallinson@ed.ac.uk sivar@stanford.edu, mlap@inf.ed.ac.uk Abstract Question answering (QA) systems are sen- sitive to the many different ways natural language expresses the same information need. In this paper we turn to paraphrases as a means of capturing this knowledge and present a general framework which learns felicitous paraphrases for various QA tasks. Our method is trained end-to- end using question-answer pairs as a su- pervision signal. A question and its para- phrases serve as input to a neural scor- ing model which assigns higher weights to linguistic expressions most likely to yield correct answers. We evaluate our approach on QA over Freebase and answer sentence selection. Experimental results on three datasets show that our framework con- sistently improves performance, achieving competitive results despite the use of sim- ple QA models. 1 Introduction Enabling computers to automatically answer ques- tions posed in natural language on any domain or topic has been the focus of much research in re- cent years. Question answering (QA) is challeng- ing due to the many different ways natural lan- guage expresses the same information need. As a result, small variations in semantically equivalent questions, may yield different answers. For exam- ple, a hypothetical QA system must recognize that the questions “who created microsoft” and “who started microsoft” have the same meaning and that they both convey the founder relation in order to retrieve the correct answer from a knowledge base. Given the great variety of surface forms for se- mantically equivalent expressions, it should come as no surprise that previous work has investigated the use of paraphrases in relation to question an- swering. There have been three main strands of research. The first one applies paraphrasing to match natural language and logical forms in the context of semantic parsing. Berant and Liang (2014) use a template-based method to heuristi- cally generate canonical text descriptions for can- didate logical forms, and then compute paraphrase scores between the generated texts and input ques- tions in order to rank the logical forms. Another strand of work uses paraphrases in the context of neural question answering models (Bordes et al., 2014a,b; Dong et al., 2015). These models are typ- ically trained on question-answer pairs, and em- ploy question paraphrases in a multi-task learning framework in an attempt to encourage the neural networks to output similar vector representations for the paraphrases. The third strand of research uses paraphrases more directly. The idea is to paraphrase the question and then submit the rewritten version to a QA module. Various resources have been used to produce question paraphrases, such as rule-based machine translation (Duboue and Chu- Carroll, 2006), lexical and phrasal rules from the Paraphrase Database (Narayan et al., …

Content truncated to 500 words