Meta Distant Transfer Learning for Pre-trained Language Models

Chengyu Wang, Haojie Pan, Minghui Qiu, Jun Huang, Fei Yang, Yin Zhang


Abstract
With the wide availability of Pre-trained Language Models (PLMs), multi-task fine-tuning across domains has been extensively applied. For tasks related to distant domains with different class label sets, PLMs may memorize non-transferable knowledge for the target domain and suffer from negative transfer. Inspired by meta-learning, we propose the Meta Distant Transfer Learning (Meta-DTL) framework to learn the cross-task knowledge for PLM-based methods. Meta-DTL first employs task representation learning to mine implicit relations among multiple tasks and classes. Based on the results, it trains a PLM-based meta-learner to capture the transferable knowledge across tasks. The weighted maximum entropy regularizers are proposed to make meta-learner more task-agnostic and unbiased. Finally, the meta-learner can be fine-tuned to fit each task with better parameter initialization. We evaluate Meta-DTL using both BERT and ALBERT on seven public datasets. Experiment results confirm the superiority of Meta-DTL as it consistently outperforms strong baselines. We find that Meta-DTL is highly effective when very few data is available for the target task.
Anthology ID:
2021.emnlp-main.768
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
9742–9752
Language:
URL:
https://aclanthology.org/2021.emnlp-main.768
DOI:
10.18653/v1/2021.emnlp-main.768
Bibkey:
Cite (ACL):
Chengyu Wang, Haojie Pan, Minghui Qiu, Jun Huang, Fei Yang, and Yin Zhang. 2021. Meta Distant Transfer Learning for Pre-trained Language Models. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 9742–9752, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Meta Distant Transfer Learning for Pre-trained Language Models (Wang et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.768.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.768.mp4
Data
IMDb Movie ReviewsMultiNLISSTSST-5