Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

Shaoyi Huang, Dongkuan Xu, Ian Yen, Yijue Wang, Sung-En Chang, Bingbing Li, Shiyang Chen, Mimi Xie, Sanguthevar Rajasekaran, Hang Liu, Caiwen Ding


Abstract
Conventional wisdom in pruning Transformer-based language models is that pruning reduces the model expressiveness and thus is more likely to underfit rather than overfit. However, under the trending pretrain-and-finetune paradigm, we postulate a counter-traditional hypothesis, that is: pruning increases the risk of overfitting when performed at the fine-tuning phase. In this paper, we aim to address the overfitting problem and improve pruning performance via progressive knowledge distillation with error-bound properties. We show for the first time that reducing the risk of overfitting can help the effectiveness of pruning under the pretrain-and-finetune paradigm. Ablation studies and experiments on the GLUE benchmark show that our method outperforms the leading competitors across different tasks.
Anthology ID:
2022.acl-long.16
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
190–200
Language:
URL:
https://aclanthology.org/2022.acl-long.16
DOI:
10.18653/v1/2022.acl-long.16
Bibkey:
Cite (ACL):
Shaoyi Huang, Dongkuan Xu, Ian Yen, Yijue Wang, Sung-En Chang, Bingbing Li, Shiyang Chen, Mimi Xie, Sanguthevar Rajasekaran, Hang Liu, and Caiwen Ding. 2022. Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 190–200, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm (Huang et al., ACL 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/naacl24-info/2022.acl-long.16.pdf
Software:
 2022.acl-long.16.software.zip
Data
GLUEQNLI