This is the source code of our method proposed in paper "Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings" accepted by COLING 2022.
Below are the path of the code to reproduce our results. Please check the README.md file in each project folder for more detailed instructions.
- sa_bert: code to pretrain SaBERT
- sa_elmo: code to pretrain SaELMo and SaELMo-Proj
Please cite our paper if you found the resources in this repository useful.
@inproceedings{liu-etal-2022-towards,
title = "Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings",
author = "Liu, Linlin and
Nguyen, Thien Hai and
Joty, Shafiq and
Bing, Lidong and
Si, Luo",
booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
month = October,
year = "2022",
address = "Gyeongju, Republic of Korea",
publisher = "International Committee on Computational Linguistics",
}