Llama 3.2 3B It Chinese Kyara

🚀 Kyara: Knowledge Yielding Adaptive Retrieval Augmentation for LLM Fine-tuning

Kyara (Knowledge Yielding Adaptive Retrieval Augmentation) is an experimental project. It aims to improve language models through knowledge retrieval processes. The project enhances the model's ability to adapt knowledge and improve language comprehension, especially for underrepresented languages like Traditional Chinese. Since there is relatively scarce Traditional Chinese data compared to the large English corpus used for model training, Kyara addresses this gap by expanding the limited corpus of this language. This is a preview model, and the stable version will be released soon.

🚀 Quick Start

This section is not provided in the original document, so it is skipped.

✨ Features

• Kyara is an experimental project focused on enhancing language models via knowledge retrieval.
• It aims to improve the model's knowledge - adaptation ability and language comprehension, especially for under - represented languages such as Traditional Chinese.
• It helps expand the limited corpus of Traditional Chinese due to the scarcity of relevant data compared to English.

📚 Documentation

Benchmark

All evaluations are conducted in a zero - shot setting.

Links

• 

• 🤗 Hugging Face  | 🚀Github  |  📃 Paper  |  📄 English  |  📄 Chinese  |  💻 Kaggle Notebook

Image

📄 License

The license for this project is llama3.2.