We create a Thai multi-turn conversation dataset from airesearch/WangchanThaiInstruct (Batch 1) by LLM. It was created from synthetic method using open source LLM in Thai language.
Thammaleelakul, S., & Phatthiyaphaibun, W. (2024). WangchanThaiInstruct Multi-turn Conversation Dataset [Data set]. Zenodo.
https://doi.org/10.5281/zenodo.13132633
or BibTeX
@dataset{thammaleelakul_2024_13132633,
author =… See the full description on the dataset page:
https://huggingface.co/datasets/ThaiSyntheticQA/WangchanThaiInstruct_Multi-turn_Conversation_Dataset.