Overview

This repository contains a medical LLM-as-a-Judge model based on Gemma-3-4B and fine-tuned for medical response evaluation tasks. This version combines evaluation modules 2 and 3 with reference answers.

Base Model

Base architecture: Gemma-3-4B
Frameworks:
- Transformers
- PEFT / LoRA
- TRL

Training Objective

The model is designed to evaluate generated medical answers according to predefined clinical evaluation criteria. Depending on the version, the model may operate:

with expert reference answers (with-ref)
without reference answers (no-ref)