A fine-tuned multilingual model for Vietnamese reasoning

NOTE

• This model is a test version for our new training pipeline. THIS IS NOT SUITABLE FOR PRODUCTION
• The full model will be updated soon.

📋 Overview

• A language model with reasoning capability that provides step-by-step reasoning in Vietnamese before delivering answers.
• The model follows a structured XML format with explicit reasoning tags.
• It's designed for educational applications and complex problem-solving tasks in Vietnamese.

🔧 Method

• Fine-tuned using Group Relative Policy Optimization (GRPO) with Unsloth for hardware efficiency.
• Employs rule-based reward functions to encourage adherence to Vietnamese XML reasoning format.
• Uses LoRA adaptation on a Vietnamese dataset spanning various task types.

💫 Quantization

🤝 Contributors