Hugging Face Overview
Korean Medical LLM by SNUH HARI
snuh/hari-q3-14b is a 14 billion parameter Large Language Model (LLM) developed by the Healthcare AI Research Institute (HARI) at Seoul National University Hospital (SNUH). This model is specifically fine-tuned for medical Question-Answer (QA) style generation, primarily in Korean and English, focusing on the clinical medicine domain.
Key Capabilities & Performance
- Medical QA Generation: Optimized for generating answers to medical questions, reflecting clinical reasoning.
- High Accuracy: Achieves 84.14% accuracy on the reasoning section of the Korean Medical Licensing Examination (KMLE) benchmark.
- Training Data: Fine-tuned on a curated corpus of publicly available, de-identified Korean medical QA data, including clinical guidelines, academic publications, and exam-style questions.
- Ethical Compliance: Trained exclusively on de-identified data, ensuring no real patient data or PII is included.
Intended Use Cases
- Clinical Decision Support: Provides QA-style assistance for medical professionals.
- Medical Education: Useful for self-assessment tools and educational platforms.
- Automated Medical Reasoning: Aids in documentation and reasoning tasks within healthcare.
Important Considerations
- This model is intended for research and educational purposes only and should not be used to make clinical decisions.
- Benchmarks are provided for research and do not imply clinical safety or efficacy.