EMNLP · 2022 · Conference paper

Improving Stability of Fine-Tuning Pretrained Language Models via Component-Wise Gradient Norm Clipping

Rendered by arXiv from the LaTeX source. If anything looks wrong, switch to the PDF.HTML · PDF · Open in a new tab ↗ · Close

Published in: Conference on Empirical Methods in Natural Language Processing
Date: 2022-10-19
Citations: 7
arXiv: 2210.10325 ↗

Cite

@inproceedings{yang2022improving,
  title = {Improving Stability of Fine-Tuning Pretrained Language Models via Component-Wise Gradient Norm Clipping},
  author = {Chenghao Yang and Xuezhe Ma},
  year = {2022},
  booktitle = {Conference on Empirical Methods in Natural Language Processing},
  eprint = {2210.10325},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2210.10325},
}