Update README.md

README.md

@@ -11,6 +11,8 @@ WARNING! Load tokenizer as AutoTokenizer.from_pretrained(model_path, use_fast=Tr
 
 Up to 60% faster generation and 35% training (on identical russian text sequences!) with HF because of different tokenizer.
 
+Paper: Tikhomirov M., Chernyshev D. Impact of Tokenization on LLaMa Russian Adaptation //arXiv preprint arXiv:2312.02598. – 2023.
+
 ## Training procedure
 
 ruadapt mistral trained on saiga corpuses.