Vietnamese Dialogue Dataset for Instruction Tuning

A dataset of short, natural Vietnamese dialogues for fine-tuning language models like Mamba, LLaMA, and Gemma. It contains everyday communication, frequently asked questions, and emotional responses, formatted as JSONL and ready for instruction tuning. The dataset was created by hoanghai2110 for the Vietnamese open-source AI community.

Use Cases

• Fine-tune a language model like LLaMA on Vietnamese dialogue for instruction tuning tasks
• Train a model to generate emotional responses based on conversational context from the dataset
• Use the dataset of everyday communication and FAQs to build a Vietnamese conversational agent

Strengths

• Formatted as JSONL, ready for instruction tuning
• Created specifically for the Vietnamese open-source AI community
• Contains everyday communication, FAQs, and emotional responses

Limitations

• Unknown dataset size, row count, and column structure
• Potential bias towards the specific conversational topics and styles collected
• No information provided on data collection methodology or validation

Provenance

Source

huggingface

Freshness

Last updated 2025-07-06.

Geography

Vietnam