NeuralREG: An end-to-end approach to referring expression generation
This work addresses referring expression generation for natural language processing applications, presenting an incremental improvement by integrating form and content decisions into a single neural model.
The paper tackled the problem of generating referring expressions by introducing NeuralREG, an end-to-end deep neural network approach that eliminates explicit feature extraction, and it showed substantial improvements over two strong baselines on the delexicalized WebNLG corpus.
Traditionally, Referring Expression Generation (REG) models first decide on the form and then on the content of references to discourse entities in text, typically relying on features such as salience and grammatical function. In this paper, we present a new approach (NeuralREG), relying on deep neural networks, which makes decisions about form and content in one go without explicit feature extraction. Using a delexicalized version of the WebNLG corpus, we show that the neural model substantially improves over two strong baselines. Data and models are publicly available.