Abstract
Traditional methods for deep NLG adopt pipeline approaches comprising stages such as constructing syntactic input, predicting function words, linearizing the syntactic input and generating the surface forms. Though easier to visualize, pipeline approaches suffer from error propagation. In addition, information available across modules cannot be leveraged by all modules. We construct a transition-based model to jointly perform linearization, function word prediction and morphological generation, which considerably improves upon the accuracy compared to a pipelined baseline system. On a standard deep input linearization shared task, our system achieves the best results reported so far.- Anthology ID:
- E17-1061
- Volume:
- Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers
- Month:
- April
- Year:
- 2017
- Address:
- Valencia, Spain
- Editors:
- Mirella Lapata, Phil Blunsom, Alexander Koller
- Venue:
- EACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 643–654
- Language:
- URL:
- https://aclanthology.org/E17-1061
- DOI:
- Cite (ACL):
- Ratish Puduppully, Yue Zhang, and Manish Shrivastava. 2017. Transition-Based Deep Input Linearization. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 643–654, Valencia, Spain. Association for Computational Linguistics.
- Cite (Informal):
- Transition-Based Deep Input Linearization (Puduppully et al., EACL 2017)
- PDF:
- https://preview.aclanthology.org/emnlp-22-attachments/E17-1061.pdf
- Code
- SUTDNLP/ZGen