Large Language Models

Developing Vietnamese LLMs and applications

Recently, large language models have achieved state-of-the-art results on many natural language processing tasks. It is believed that large language models can learn the knowledge of the world from the text data. As the era of AGI (Artificial General Intelligence) is coming, large language models are considered as one of the key technologies to achieve AGI.

Key Projects

URA-LLaMa Family

Fine-tuned Vietnamese LLMs from Meta’s LLaMa-2 model, including 7B, 13B, and 70B versions.

MixSUra & GemSUra

Vietnamese language models based on Mixtral and Gemma architectures.

Vision-Language Models

Integrating vision capabilities with LLMs using the LLaVA architecture:

  • MixSUraV
  • GemSUraV 7B, 2B

Research Interests

  • Developing new large language models for low-resource languages
  • Applying LLMs to solve real-world problems
  • Cross-lingual transfer learning

Citations

If you use our models in your research, please cite this paper (Truong et al., 2024).

References

2024

  1. Crossing Linguistic Horizons: Finetuning and Comprehensive Evaluation of Vietnamese Large Language Models
    Sang Truong, Duc Nguyen, Toan Nguyen, and 4 more authors
    In Findings of the Association for Computational Linguistics: NAACL 2024, Jun 2024