T-LLaMA / README.md
Pagewood's picture
Update README.md
d1771e2 verified
|
raw
history blame
332 Bytes
metadata
language:
  - bo

T‑LLaMA: a Tibetan large language model based on LLaMA2

In this study, we trained Tibetan LLaMA based on LLaMA2 and achieved state-of-the-art performance in the text classification task, with an accuracy of 79.8%. Additionally, we obtained promising results in text generation and text summarization tasks.