"HuatuoGPT-3-27B is a medical LLM built on Qwen3.8-27B with One-stage Policy Optimization (OnePO). OnePO adapts language models to medicine in a single reinforcement-learning stage, without preceding domain-specific supervised fine-tuning. Teacher responses provide temporary guidance and are retired as the model improves.

We release the training code, medical RL dataset, and 8B rubric grader."

**link to the paper (pdf): https://openreview.net/pdf?id=M8eyUQldfx

im skipping the AI summary for this one… because: 1) primarily i dont have time 2) i didnt choose this paper soley for unique methodology 3) other reasons