Abstract
We address the problem of model generalization for sequence to sequence (seq2seq) architectures. We propose going beyond data augmentation via paraphrase-optimized multi-task learning and observe that it is useful in correctly handling unseen sentential paraphrases as inputs. Our models greatly outperform SOTA seq2seq models for semantic parsing on diverse domains (Overnight - up to 3.2% and emrQA - 7%) and Nematus, the winning solution for WMT 2017, for Czech to English translation (CzENG 1.6 - 1.5 BLEU).- Anthology ID:
- 2020.clinicalnlp-1.30
- Volume:
- Proceedings of the 3rd Clinical Natural Language Processing Workshop
- Month:
- November
- Year:
- 2020
- Address:
- Online
- Editors:
- Anna Rumshisky, Kirk Roberts, Steven Bethard, Tristan Naumann
- Venue:
- ClinicalNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 269–279
- Language:
- URL:
- https://aclanthology.org/2020.clinicalnlp-1.30
- DOI:
- 10.18653/v1/2020.clinicalnlp-1.30
- Cite (ACL):
- So Yeon Min, Preethi Raghavan, and Peter Szolovits. 2020. Advancing Seq2seq with Joint Paraphrase Learning. In Proceedings of the 3rd Clinical Natural Language Processing Workshop, pages 269–279, Online. Association for Computational Linguistics.
- Cite (Informal):
- Advancing Seq2seq with Joint Paraphrase Learning (Min et al., ClinicalNLP 2020)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2020.clinicalnlp-1.30.pdf
- Data
- emrQA