tykea/mBart-large-50-KQA
Text2Text Generation
•
Updated
•
31
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.