Update README.md

README.md

@@ -16,13 +16,12 @@ library_name: transformers
 
 ### Model Description
 
-GemmaX2-28-9B-Pretrain is a language model that results from continual pretraining of Gemma2-9B on a mix of 56 billion tokens of monolingual and parallel data in 28 different languages — Arabic, Bengali, Czech, German, English, Spanish, Persian, French, Hebrew, Hindi, Indonesian, Italian, Japanese, Khmer, Korean, Lao, Malay, Burmese, Dutch, polish, Portuguese, Russian, Thai, Tagalog, Turkish, Urdu, Vietnamese, Chinese. 
+GemmaX2-28-9B-v0.1 is an LLM-based translation model. It has been fintuned on GemmaX2-28-9B-Pretrain, which is a language model developed through continual pretraining of Gemma2-9B using a mix of 56 billion tokens from both monolingual and parallel data across 28 different languages. Please find more details in our paper: [Multilingual Machine Translation with Open Large Language Models at Practical Scale: An Empirical Study](https://arxiv.org/abs/2502.02481).
 
-GemmaX2-28-9B-v0.1 is the first model in the series. Compared to the current open-source state-of-the-art (SOTA) models, it achieves optimal translation performance across 28 languages, even reaching performance comparable to GPT-4 and Google Translate, indicating it has achieved translation capabilities on par with industry standards. 
 
 - **Developed by:** Xiaomi
-- **Model type:** A 9B parameter model base on Gemma2, we obtained GemmaX2-28-9B-Pretrain by continuing pre-training on a large amount of monolingual and parallel data. Afterward, GemmaX2-28-9B-v0.1 was derived through supervised fine-tuning on a small set of high-quality instruction data.
-- **Language(s) (NLP):** Arabic, Bengali, Czech, German, English, Spanish, Persian, French, Hebrew, Hindi, Indonesian, Italian, Japanese, Khmer, Korean, Lao, Malay, Burmese, Dutch, polish, Portuguese, Russian, Thai, Tagalog, Turkish, Urdu, Vietnamese, Chinese. 
+- **Model type:** GemmaX2-28-9B-Pretrain is obtained by continually pretraining Gemma2-9B on a large amount of monolingual and parallel data. Subsequently, GemmaX2-28-9B-v0.1 is derived through supervised finetuning on a small set of high-quality translation instruction data.
+- **Languages:** Arabic, Bengali, Czech, German, English, Spanish, Persian, French, Hebrew, Hindi, Indonesian, Italian, Japanese, Khmer, Korean, Lao, Malay, Burmese, Dutch, polish, Portuguese, Russian, Thai, Tagalog, Turkish, Urdu, Vietnamese, Chinese. 
 - **License:**  gemma
 
 ### Model Source
@@ -33,10 +32,6 @@ GemmaX2-28-9B-v0.1 is the first model in the series. Compared to the current ope
 
 ![Experimental Result](main.png)
 
-## Limitations
-
-GemmaX2-28-9B-v0.1 supports only the 28 most commonly used languages and does not guarantee powerful translation performance for other languages. Additionally, we will continue to improve GemmaX2-28-9B's translation performance, and future models will be release in due course.
-
 
 
 ## Run the model
@@ -56,9 +51,6 @@ outputs = model.generate(**inputs, max_new_tokens=50)
 print(tokenizer.decode(outputs[0], skip_special_tokens=True))
 ```
 
-### Training Data
-
-We collected monolingual data from [CulturaX](https://huggingface.co/datasets/uonlp/CulturaX) and [MADLAD-400](https://huggingface.co/datasets/allenai/MADLAD-400). For parallel data, we collected all Chinese-centric and English-centric parallel dataset from the [OPUS](https://opus.nlpl.eu/) collection up to Auguest 2024 and underwent a series of filtering processes, such as language detection, semantic duplication filtering, quality filtering, and more.
 
 ## Citation 
 
@@ -72,4 +64,9 @@ We collected monolingual data from [CulturaX](https://huggingface.co/datasets/uo
       primaryClass={cs.CL},
       url={https://arxiv.org/abs/2502.02481}, 
 }
-```
+```
+
+
+## Limitations
+
+GemmaX2-28-9B-v0.1 supports only the 28 most commonly used languages and does not guarantee powerful translation performance for other languages. Additionally, we will continue to improve GemmaX2-28-9B's translation performance, and future models will be release in due course.