Year: 2018
Type: conference-paper
Abstract: Text in many domains involves a significant amount of named entities.Predicting the entity names is often challenging for a language model as they appear less frequent on the training corpus.In this paper, we propose a novel and effective approach to building a discriminative language model which can learn the entity names by leveraging their entity type information.We also introduce two benchmark datasets based on recipes and Java programming codes, on which we evaluate the proposed model.Experimental results show that our model achieves 52.2% better perplexity in recipe generation and 22.06% on code generation than the stateof-the-art language models.
DOI: 10.18653/v1/p18-1221
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Funders: National Science Foundation, Nvidia
OpenAlex ID: W2963676655
