The UPC RDF-to-Text System at WebNLG Challenge 2020

View/Open
Cita com:
hdl:2117/366164
Document typeConference report
Defense date2020
PublisherAssociation for Computational Linguistics
Rights accessOpen Access
This work is protected by the corresponding intellectual and industrial property rights.
Except where otherwise noted, its contents are licensed under a Creative Commons license
:
Attribution 3.0 Spain
Abstract
This work describes the end-to-end system architecture presented at WebNLG Challenge 2020. The system follows the traditional Machine Translation (MT) pipeline, based on the Transformer model, applied in most text-to-text problems. Our solution is enriched by means of a Back Translation step over the original corpus. Thus, the system directly relies on lexicalise format since the synthetic data limits the use of delexicalisation.
CitationBergés, D. [et al.]. The UPC RDF-to-Text System at WebNLG Challenge 2020. A: WebNLG - International Workshop on Natural Language Generation from the Semantic Web. "Proceedings of the 3rd International Workshop on Natural Language Generation from the Semantic Web (WebNLG+)". Stroudsburg, PA: Association for Computational Linguistics, 2020, p. 167-170.
Publisher versionhttps://aclanthology.org/2020.webnlg-1.19/
Files | Description | Size | Format | View |
---|---|---|---|---|
2020.webnlg-1.19.pdf | 385,2Kb | View/Open |