haoranxu commited on
Commit
5abb78d
1 Parent(s): 3c93f68

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -2
README.md CHANGED
@@ -3,10 +3,21 @@ language:
3
  - en
4
  - de
5
  ---
6
- Our bibert-ende is a bilingual English-German Language Model.
 
 
 
 
 
 
 
 
 
 
7
  # Download
 
8
  ```
9
  from transformers import BertTokenizer, AutoModel
10
  tokenizer = BertTokenizer.from_pretrained("jhu-clsp/bibert-ende")
11
  model = AutoModel.from_pretrained("jhu-clsp/bibert-ende")
12
- ```
 
3
  - en
4
  - de
5
  ---
6
+ Our bibert-ende is a bilingual English-German Language Model. Please check out our EMNLP 2021 paper "[BERT, mBERT, or BiBERT? A Study on Contextualized Embeddings for Neural Machine Translation](https://arxiv.org/abs/2109.04588)" for more details.
7
+ ```
8
+ @misc{xu2021bert,
9
+ title={BERT, mBERT, or BiBERT? A Study on Contextualized Embeddings for Neural Machine Translation},
10
+ author={Haoran Xu and Benjamin Van Durme and Kenton Murray},
11
+ year={2021},
12
+ eprint={2109.04588},
13
+ archivePrefix={arXiv},
14
+ primaryClass={cs.CL}
15
+ }
16
+ ```
17
  # Download
18
+
19
  ```
20
  from transformers import BertTokenizer, AutoModel
21
  tokenizer = BertTokenizer.from_pretrained("jhu-clsp/bibert-ende")
22
  model = AutoModel.from_pretrained("jhu-clsp/bibert-ende")
23
+ ```