Model Overview

MultiClinNER-UniboNLP/medgemma-en-ner-en-disease-3epochs-clean is a specialized 4.3 billion parameter model developed by MultiClinNER-UniboNLP. It is fine-tuned from the unsloth/medgemma-4b-it-bnb-4bit base model, leveraging the Gemma3 architecture. This model was specifically trained for English Named Entity Recognition (NER) with a focus on identifying disease entities within text.

Key Capabilities

• Disease Named Entity Recognition: Excels at identifying and extracting disease names and related medical conditions from English text.
• Efficient Fine-tuning: Benefits from being trained 2x faster using Unsloth and Huggingface's TRL library, indicating an optimized training process.
• Gemma3 Architecture: Built upon the Gemma3 model family, providing a robust foundation for its NER capabilities.

Good For

• Medical Text Analysis: Ideal for applications requiring the extraction of disease information from clinical notes, research papers, or other medical documents.
• Biomedical Information Extraction: Useful in scenarios where precise identification of disease entities is crucial for downstream tasks like knowledge graph construction, clinical decision support, or epidemiological studies.
• Research and Development: Provides a specialized tool for researchers working on medical NLP tasks, particularly those involving disease entity recognition.