BAM! Born-Again Multi-Task Networks for Natural Language Understanding

Kevin Clark, Minh-Thang Luong, Urvashi Khandelwal, Christopher D. Manning, Quoc V. Le


Abstract
It can be challenging to train multi-task neural networks that outperform or even match their single-task counterparts. To help address this, we propose using knowledge distillation where single-task models teach a multi-task model. We enhance this training with teacher annealing, a novel method that gradually transitions the model from distillation to supervised learning, helping the multi-task model surpass its single-task teachers. We evaluate our approach by multi-task fine-tuning BERT on the GLUE benchmark. Our method consistently improves over standard single-task and multi-task training.
Anthology ID:
P19-1595
Volume:
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2019
Address:
Florence, Italy
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5931–5937
Language:
URL:
https://aclanthology.org/P19-1595
DOI:
10.18653/v1/P19-1595
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/P19-1595.pdf
Video:
 https://vimeo.com/385244777
Data
CoLAGLUEMRPCSST