Building applied natural language generation systems

By Ehud Reiter, Rick Dale. 1997

In this article, we give an overview of Natural Language Generation (NLG) from an applied system-building perspective. The article includes a discussion of when NLG techniques should be used; suggestions for carrying out requirements analyses; and a description of the basic NLG tasks of content determination, discourse planning, sentence aggregation, lexicalization, referring expression generation, and linguistic realisation. Throughout, the emphasis is on established techniques that can be used to build simple but practical working systems now. We also provide pointers to techniques in the literature that are appropriate for more complicated scenarios.

Subfields: 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 Natural Language Processing

Citations in the corpus

Survey of Hallucination in Natural Language Generation (2022, background): “The source in abstractive summarization is the input source text that is being summarized [165], while the source in data-totext is non-linguistic data [56, 157], and the source(s) in the dialogue system is dialogue history and/or the external knowledge sentences.”
Challenges in Data-to-Document Generation (2017, background): “A classic problem in natural-language generation (NLG) (Kukich, 1983; McKeown, 1992; Reiter and Dale, 1997) involves taking structured data, such as a table, as input, and producing text”
ToTTo: A Controlled Table-To-Text Generation Dataset (2020, background): “Prior to the advent of neural approaches, generation systems typically separated content selection ( what to say ) from surface realization ( how to say it ) (Reiter and Dale, 1997).”
Handling Divergent Reference Texts when Evaluating Table-to-Text Generation (2019, background): “The task of generating natural language descriptions of structured data (such as tables) (Kukich, 1983; McKeown, 1985; Reiter and Dale, 1997) has seen a growth in interest with the rise of sequence to sequence models that provide an easy way of encoding tables and generating text from them (Lebret…”
Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation (2017, methodology): “This characterisation suggests a close similarity to lexicalisation, but Reiter and Dale (2000) point out that the essential difference is that referring expression generation is a ‘discrimination task, where the system needs to communicate sufficient information to distinguish one domain entity fro”
Barack’s Wife Hillary: Using Knowledge Graphs for Fact-Aware Language Modeling (2019, methodology): “For a survey of early non-neural text generation meth-ods we refer the reader to Reiter and Dale (1997).”
Evaluation of Text Generation: A Survey (2020, background): “In their seminal paper, Reiter & Dale (2000) present a classical three-stage NLG architecture, as shown in Figure 1.1.”
Few-Shot NLG with Pre-Trained Language Model (2019, methodology): “Such a pipeline-based paradigm (Reiter and Dale 1997) divides the generation process into parts with separate and explicit objectives such as content selection followed by macro/micro-planning and surface realization.”
Data-to-Text Generation with Content Selection and Planning (2018, background): “The generation literature provides multiple examples of content selection components developed for various domains which are either hand-built (Kukich 1983; McKeown 1992; Reiter and Dale 1997; Duboue and McKeown 2003) or learned from data (Barzilay and Lapata 2005; Duboue and McKeown 2001; 2003; Lia”
Planning with Learned Entity Prompts for Abstractive Summarization (2021, background): “By planning the whole summary beforehand, the summary-level planner would be (i) less susceptible to local correlations than the sentence-level planning while generating the entities, and (ii) a better microplanner in deciding sentence boundaries and avoiding verbose summaries (Reiter and Dale, 1997”
A Survey of the State of Explainable AI for Natural Language Processing (2020, background): “In fact, many declarative induction-based techniques can use template-based natural language generation (Reiter and Dale, 1997) to turn rules and programs into human-comprehensible language, and this minor extension can potentially make the explanation more accessible to lay users.”
Discriminative Syntax-Based Word Ordering for Text Generation (2015, unknown): “Word ordering is a fundamental problem in natural language generation (NLG, Reiter and Dale 1997).”
Text Generation with Efficient (Soft) Q-Learning (2021, methodology): “To demonstrate this, we consider the task of prompting a large pretrained LM [47, 4] for controllable generation [51, 20].”
Good Question! Statistical Ranking for Question Generation (2010, background): “QG is very different from many natural language generation problems because the input is natural language rather than a formal representation (cf. Reiter and Dale, 1997).”
Neural Sentence Ordering (2016, background): “The task is general and yet challenging, and is especially important for natural language generation (Reiter and Dale, 1997).”
Denoising Pre-Training and Data Augmentation Strategies for Enhanced RDF Verbalization with Transformers (2020, unknown): “Following the traditional NLG pipeline ( Reiter and Dale, 1997), early works on factual data ver-balization mainly focused on the creation of rules or templates to produce the textual output.”