Natural Language Processing
Published November 1, 2019 | Version 1.0
Dataset Open

Transfer fine-tuned BERT models by paraphrases

Creators

  • 1. Osaka University
  • 2. Artificial Intelligence Research Center (AIRC), AIST

Description

Transfer fine-tuned BERT models by phrasal paraphrases. 

  • transferFT_bert-base-uncased.pkl bases on the bert-base-uncased model
  • transferFT_bert-large-uncased.pkl bases on the bert-large-uncased model

For usage, please refer to our GitHub page.

https://github.com/yukiar/TransferFT

For details of these models, please refer to our paper.

Yuki Arase and Junichi Tsujii. 2019. Transfer Fine-Tuning: A BERT Case Study. in Proc. of Conference on Empirical Methods in Natural Language Processing (EMNLP 2019).

https://arxiv.org/abs/1909.00931

Notes

When you have publications using our models, please cite our EMNLP2019 paper.

Files

Files (1.8 GB)

Name Size Download all
md5:5bb112fb21e6006c538074056336cfc0
438.0 MB Download
md5:e22a668fb03289c9c360f2448117e05e
1.3 GB Download