Semi-Supervised QA with Generative Domain-Adaptive Nets

Zhilin Yang, Junjie Hu, Ruslan Salakhutdinov, William Cohen


Abstract
We study the problem of semi-supervised question answering—utilizing unlabeled text to boost the performance of question answering models. We propose a novel training framework, the Generative Domain-Adaptive Nets. In this framework, we train a generative model to generate questions based on the unlabeled text, and combine model-generated questions with human-generated questions for training question answering models. We develop novel domain adaptation algorithms, based on reinforcement learning, to alleviate the discrepancy between the model-generated data distribution and the human-generated data distribution. Experiments show that our proposed framework obtains substantial improvement from unlabeled text.
Anthology ID:
P17-1096
Volume:
Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2017
Address:
Vancouver, Canada
Editors:
Regina Barzilay, Min-Yen Kan
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1040–1050
Language:
URL:
https://aclanthology.org/P17-1096
DOI:
10.18653/v1/P17-1096
Bibkey:
Cite (ACL):
Zhilin Yang, Junjie Hu, Ruslan Salakhutdinov, and William Cohen. 2017. Semi-Supervised QA with Generative Domain-Adaptive Nets. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1040–1050, Vancouver, Canada. Association for Computational Linguistics.
Cite (Informal):
Semi-Supervised QA with Generative Domain-Adaptive Nets (Yang et al., ACL 2017)
Copy Citation:
PDF:
https://aclanthology.org/P17-1096.pdf
Video:
 https://aclanthology.org/P17-1096.mp4
Data
SQuAD