# Stabilizing Reasoning in Medical LLMs with Continued Pretraining and Reasoning Preference Optimization

arXiv: https://arxiv.org/abs/2504.18080 · HTML: https://mlsanity.com/paper/2504.18080/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 29 of its 41 references that are in the map.
- First posted on arXiv 25 Apr 2025.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)
- LLM alignment / RLHF / preference optimisation / [Medical LLM alignment](https://mlsanity.com/alignment/medical-llm-alignment/)

## Builds on (most cited first, 29 of 29)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [MedMCQA : A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answering](https://mlsanity.com/paper/2203.14371/) (2022)
- [Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine](https://mlsanity.com/paper/2311.16452/) (2023)
- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs](https://mlsanity.com/paper/2412.18925/) (2024)
- [The case for 4-bit precision: k-bit Inference Scaling Laws](https://mlsanity.com/paper/2212.09720/) (2022)
- [Iterative Reasoning Preference Optimization](https://mlsanity.com/paper/2404.19733/) (2024)
- [ReFT: Reasoning with Reinforced Fine-Tuning](https://mlsanity.com/paper/2401.08967/) (2024)
- [Capabilities of Gemini Models in Medicine](https://mlsanity.com/paper/2404.18416/) (2024)
- [Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?](https://mlsanity.com/paper/2405.05904/) (2024)
- [HuatuoGPT, towards Taming Language Model to Be a Doctor](https://mlsanity.com/paper/2305.15075/) (2023)
- [Medical Hallucinations in Foundation Models and Their Impact on Healthcare](https://mlsanity.com/paper/2503.05777/) (2025)
- [Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing Examinations](https://mlsanity.com/paper/2303.18027/) (2023)
- [A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor?](https://mlsanity.com/paper/2409.15277/) (2024)
- [Superhuman performance of a large language model on the reasoning tasks of a physician](https://mlsanity.com/paper/2412.10849/) (2024)
- [Towards a Personal Health Large Language Model](https://mlsanity.com/paper/2406.06474/) (2024)
- [MEDIC: Comprehensive Evaluation of Leading Indicators for LLM Safety and Utility in Clinical Applications](https://mlsanity.com/paper/2409.07314/) (2024)
- [From Medprompt to o1: Exploration of Run-Time Strategies for Medical Challenge Problems and Beyond](https://mlsanity.com/paper/2411.03590/) (2024)
- [JMedBench: A Benchmark for Evaluating Japanese Biomedical Large Language Models](https://mlsanity.com/paper/2409.13317/) (2024)
- [70B-parameter large language models in Japanese medical question-answering](https://mlsanity.com/paper/2406.14882/) (2024)
- [Development and bilingual evaluation of Japanese medical large language model within reasonably low computational resources](https://mlsanity.com/paper/2409.11783/) (2024)
- [PLaMo-100B: A Ground-Up Language Model Designed for Japanese Proficiency](https://mlsanity.com/paper/2410.07563/) (2024)

## Built on by (most cited first, 4 of 4)

- [The Sound of Syntax: Finetuning and Comprehensive Evaluation of Language Models for Speech Pathology](https://mlsanity.com/paper/2509.16765/) (2025)
- [MedRECT: A Medical Reasoning Benchmark for Error Correction in Clinical Texts](https://mlsanity.com/paper/2511.00421/) (2025)
- [Finding and Reactivating Post-Trained LLMs' Hidden Safety Mechanisms](https://mlsanity.com/paper/2604.00012/) (2026)
- [Computational KJ-Ho: An Analyst-Bias-Free Insight Extraction Framework from Large-Scale Qualitative Data Using Domain-Specialized LLMs](https://mlsanity.com/paper/2608.16467/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) show potential in medicine, yet clinical adoption is hindered by concerns over factual accuracy, language-specific limitations (e.g., Japanese), and critically, their reliability when required to generate reasoning explanations -- a prerequisite for trust. This paper introduces Preferred-MedLLM-Qwen-72B, a 72B-parameter model optimized for the Japanese medical domain to achieve both high accuracy and stable reasoning. We employ a two-stage fine-tuning process on the Qwen2.5-72B base model: first, Continued Pretraining (CPT) on a comprehensive Japanese medical corpus instills deep domain knowledge. Second, Reasoning Preference Optimization (RPO), a preference-based method, enhances the generation of reliable reasoning pathways while preserving high answer accuracy. Evaluations on the Japanese Medical Licensing Exam benchmark (IgakuQA) show Preferred-MedLLM-Qwen-72B achieves state-of-the-art performance (0.868 accuracy), surpassing strong proprietary models like GPT-4o (0.866). Crucially, unlike baseline or CPT-only models which exhibit significant accuracy degradation (up to 11.5% and 3.8% respectively on IgakuQA) when prompted for explanations, our model maintains its high accuracy (0.868) under such conditions. This highlights RPO's effectiveness in stabilizing reasoning generation. This work underscores the importance of optimizing for reliable explanations alongside accuracy. We release the Preferred-MedLLM-Qwen-72B model weights to foster research into trustworthy LLMs for specialized, high-stakes applications.