English

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

Computation and Language 2023-01-18 v4

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.

Keywords

Cite

@article{arxiv.2110.08190,
  title  = {Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm},
  author = {Shaoyi Huang and Dongkuan Xu and Ian E. H. Yen and Yijue Wang and Sung-en Chang and Bingbing Li and Shiyang Chen and Mimi Xie and Sanguthevar Rajasekaran and Hang Liu and Caiwen Ding},
  journal= {arXiv preprint arXiv:2110.08190},
  year   = {2023}
}

Comments

11 pages; 16 figures; Published in Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing

R2 v1 2026-06-24T06:55:30.991Z