MediX-R1: Open Ended Medical Reinforcement Learning

TL;DR

MediX-R1 introduces a novel open-ended reinforcement learning framework for medical multimodal large language models, enabling clinically grounded, free-form reasoning and surpassing existing baselines with a comprehensive reward and evaluation system.

Contribution

It presents a new RL approach with multi-signal rewards and LLM-based evaluation for medical multimodal models, improving open-ended reasoning capabilities.

Findings

Outperforms strong open-source medical LLM and VLM benchmarks.

Achieves significant gains on open-ended clinical tasks.

Effective with only ~51K instruction examples.

Abstract

We introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical multimodal large language models (MLLMs) that enables clinically grounded, free-form answers beyond multiple-choice formats. MediX-R1 fine-tunes a baseline vision-language backbone with Group Based RL and a composite reward tailored for medical reasoning: an LLM-based accuracy reward that judges semantic correctness with a strict YES/NO decision, a medical embedding-based semantic reward to capture paraphrases and terminology variants, and lightweight format and modality rewards that enforce interpretable reasoning and modality recognition. This multi-signal design provides stable, informative feedback for open-ended outputs where traditional verifiable or MCQ-only rewards fall short. To measure progress, we propose a unified evaluation framework for both text-only and image+text tasks that uses a Reference-based LLM-as-judge in place of brittle string-overlap metrics, capturing semantic correctness, reasoning, and contextual alignment. Despite using only $\sim51$K instruction examples, MediX-R1 achieves excellent results across standard medical LLM (text-only) and VLM (image + text) benchmarks, outperforming strong open-source baselines and delivering particularly large gains on open-ended clinical tasks. Our results demonstrate that open-ended RL with comprehensive reward signals and LLM-based evaluation is a practical path toward reliable medical reasoning in multimodal models. Our trained models, curated datasets and source code are available at https://medix.cvmbzuai.com