Text production has many applications. It is used, for instance, to generate dialogue turns from dialogue moves, verbalise the content of knowledge bases, or generate English sentences from rich linguistic representations, such as dependency trees or abstract meaning representations. Text production is also at work in text-to-text transformations such as sentence compression, sentence fusion, paraphrasing, sentence (or text) simplification, and text summarisation. This book offers an overview of the fundamentals of neural models for text production. In particular, we elaborate on three main aspects of neural approaches to text production: how sequential decoders learn to generate adequate text, how encoders learn to produce better input representations, and how neural generators account for task-specific objectives. Indeed, each text-production task raises a slightly different challenge (e.g, how to take the dialogue context into account when producing a dialogue turn, how to detect and merge relevant information when summarising a text, or how to produce a well-formed text that correctly captures the information contained in some input data in the case of data-to-text generation). We outline the constraints specific to some of these tasks and examine how existing neural models account for them. More generally, this book considers text-to-text, meaning-to-text, and data-to-text transformations. It aims to provide the audience with a basic knowledge of neural approaches to text production and a roadmap to get them started with the related work. The book is mainly targeted at researchers, graduate students, and industrials interested in text production from different forms of inputs.

Autorentext

Shashi Narayan is a research scientist at Google in London. Prior to joining Google, he was a post-doctoral researcher at the University of Edinburgh. This book was written while he was still at the University of Edinburgh. He received his Ph.D. from the University of Lorraine. His research focuses on natural language generation understanding and structured predictions. The questions raised in his research are relevant to various natural language applications such as question answering, paraphrase generation, semantic and syntactic parsing, document understanding and summarisation, and text simplification. His research has appeared in computational linguistics journals (e.g., TACL, Computational Linguistics,and Pattern Recognition Letters) and in conferences (e.g., ACL, EMNLP, NAACL, COLING, EACL and INLG). He was nominated to the SIGGEN board (2012-14) as a student member. He co-organised the WebNLG Shared Task, a challenge on generating text from RDF data. He served as an area co-chair for Generation at NAACL HLT 2018 and ACL 2020, and for Summarisation at ACL and EMNLP 2019.Claire Gardent is a research scientist at CNRS, the French National Center for Scientific Research. Prior to joining the CNRS, she worked at the Universite de Clermont-Ferrand (France), Saarbrucken Universitat (Germany), Utrecht, and Amsterdam Universiteit (The Netherlands). She received her Ph.D. from the University of Edinburgh and her M.Sc. from Essex University. She was nominated Chair of the EACL and acted as program chair for various international conferences, workshops, and summer schools (EACL, ENLG, SemDIAL, SIGDIAL, ESSLLI,*SEM). She served on the editorial board of the journals Computational Linguistics, Journal of Semantics and Traitement Automatique des Langues, recently headed the WebNLG project (Nancy,Bolzano, Stanford SRI), and acted as chair of SIGGEN, the ACL Special Interest Group in Natural Language Generation. She also co-organised the WebNLG Shared Task, a challenge on generating text from RDF data. Her research interests include executable semantic parsing, natural language generation, question answering, dialogue and the use of computational linguistics for linguistic analysis.