README.md

metadata

license: mit
task_categories:
  - question-answering
  - text-generation
language:
  - km
pretty_name: Khmer
size_categories:
  - n<1K

The data collected from https://www.khsearch.com/ related to the general question-answering examination. It used to train fine-tuned models from many LLMs, including LlaMa, Qwen, Mistral, and Gemma. Under the research title "Fine-tuning for Question Answering in Low-Resource Languages: A Case Study on Khmer" conducted at ViLa Lab, Institute of Technology of Cambodia, Phnom Penh.

• Lab Info: https://www.facebook.com/vilalabitc
• Paper will availble on online soon