Update README.md

README.md

@@ -29,6 +29,7 @@ SambaLingo-Bulgarian-Base is a pretrained Bi-lingual Bulgarian and English model
 - **Language(s):** Bulgarian, English
 - **Finetuned from model:** [Llama 2](https://huggingface.co/meta-llama/Llama-2-7b-hf)
 - **Try the chat version of this model**: [SambaLingo-chat-space](https://huggingface.co/spaces/sambanovasystems/SambaLingo-chat-space).
+- **Paper:** [SambaLingo: Teaching Large Language Models New Languages](https://arxiv.org/abs/2404.05829)
 - **Blog Post**: [sambalingo-open-source-language-experts](https://sambanova.ai/blog/sambalingo-open-source-language-experts)
 
 ## Getting Started
@@ -54,16 +55,7 @@ All pre-training is done on the [Cultura-X](https://huggingface.co/datasets/uonl
 We extended the vocabulary of the base llama model from 32,000 tokens to 57,000 tokens by adding up to 25,000 non-overlapping tokens from the new language.
 
 ## Evaluation
-|| SambaLingo-Bulgarian-Base     | mGPT-1.3B-bulgarian | bloom-7b1 | xglm-7.5B | mGPT-13B |      
-|-------------------------------|---------------------|-----------|-----------|----------|--------|
-| Perplexity (Lower Is Better)  | **1.416**               | 1.755     | 2.051     | 1.502    | 1.889  |
-| FLORES en->bg (8 shot, CHRF)  | **0.558**               | 0.143     | 0.211     | 0.484    | 0.136  |
-| FLORES bg->en (8 shot, CHRF)  | **0.621**               | 0.227     | 0.182     | 0.347    | 0.145  |
-| FLORES en->bg  (8 shot, BLEU) | **0.275**               | 0.006     | 0.006     | 0.093    | 0.005  |
-| FLORES bg->en  (8 shot, BLEU) | **0.340**               | 0.024     | 0.018     | 0.210    | 0.006  |
-| Belebele (3 shot)             | **48.56%**              | 21.22%    | 24.22%    | 25.89%   | 25.33% |
-| SIB-200 (3 shot)              | 61.27%              | 42.65%    | 37.25%    | **64.71%**   | 41.67% |
-| XNLI (0 shot)                 | 42.83%              | 33.75%    | 39.38%    | 44.91%   | **45.83%** |
+For evaluation results see our paper: [SambaLingo: Teaching Large Language Models New Languages](https://arxiv.org/abs/2404.05829)
 
 ## Uses
 <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
@@ -107,12 +99,12 @@ We would like to give a special thanks to the following groups:
 
 ## Cite SambaLingo
 ```
-@software{sambalingo,
-  title = {{SambaLingo: Open Source Langauge Experts}},
-  author = {SambaNova Systems},
-  url = {https://huggingface.co/sambanovasystems/SambaLingo-Bulgarian-Base}
-  month = {2},
-  year = {2024},
-  version = {1.0},
+@misc{csaki2024sambalingo,
+      title={SambaLingo: Teaching Large Language Models New Languages}, 
+      author={Zoltan Csaki and Bo Li and Jonathan Li and Qiantong Xu and Pian Pawakapan and Leon Zhang and Yun Du and Hengyu Zhao and Changran Hu and Urmish Thakker},
+      year={2024},
+      eprint={2404.05829},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
 }
 ```