nirasha/qwen2.5-7b-fact-inclusion-lora

Qwen2.5-7B Fact Inclusion LoRA

Task

Given:

• Summary triple
• Supporting text triples

The model predicts:

• is_supported
• supporting evidence
• confidence
• reasoning

Training

• Framework: Unsloth + PEFT

• Method: LoRA fine-tuning

• Dataset: Custom triplet-based fact inclusion annotations

• Developed by: nirasha

• License: apache-2.0

• Finetuned from model : unsloth/qwen2.5-7b-unsloth-bnb-4bit

This qwen2 model was trained 2x faster with Unsloth