Name
Abstract | ||
|---|---|---|
End-to-end neural approaches have achieved state-of-the-art performance in many natural language processing (NLP) tasks. Yet, they often lack transparency of the underlying decision-making process, hindering error analysis and certain model improvements. In this work, we revisit the binary linearization approach to surface realization, which exhibits more interpretable behavior, but was falling short in terms of prediction accuracy. We show how enriching the training data to better capture word order constraints almost doubles the performance of the system. We further demonstrate that encoding both local and global prediction contexts yields another considerable performance boost. With the proposed modifications , the system which ranked low in the latest shared task on multilingual surface realization now achieves best results in five out of ten languages, while being on par with the state-of-the-art approaches in others. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.18653/v1/w19-8635 | INLG |
Field | DocType | Citations |
Training set,Transparency (graphic),Word order,Ranking,Computer science,Artificial intelligence,Natural language processing,Machine learning,Linearization,Binary number,Encoding (memory) | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yevgeniy Puzikov | 1 | 2 | 0.71 |
| Claire Gardent | 2 | 0 | 0.34 |
| Ido Dagan | 3 | 4057 | 554.50 |
| Iryna Gurevych | 4 | 2471 | 189.26 |