Add reference paper

README.md

@@ -1,5 +1,5 @@
 # Twitter-XLM-Roberta-base
-This is a XLM-Roberta-base model trained on ~198M multilingual tweets, described and evaluated in the [TODO: ADD PAPER](https://example.com). To evaluate this and other LMs on Twitter-specific data, please refer to the [TODO: ADD REPO LINK](https://example.com).
+This is a XLM-Roberta-base model trained on ~198M multilingual tweets, described and evaluated in the [reference paper](https://arxiv.org/abs/2104.12250). To evaluate this and other LMs on Twitter-specific data, please refer to the [main repository](https://github.com/cardiffnlp/xlm-t).
 
 ## Preprocess Text 
 Replace usernames and links for placeholders: "@user" and "http".