Improving Heart-Focused Medical Question Answering in LLMs via Variance-Aware Rubric Rewards with GRPO

arXiv:2606.05174 · cs.CL, cs.AI · Submitted 2026-04-17 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-04-17

Updated: 2026-10-13

Comments: 29 pages main text plus 30 pages Supplementary Information

Code: https://github.com/INQUIRELAB/variance-aware-rubric-rewards-grpo

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Aggregation — 3× in this paper
• GRPO — 3× in this paper · explained in 6 episodes
• Accuracy — 2× in this paper · explained in 1 episode
• Group Relative Policy Optimization — 1× in this paper
• Group Relative Policy Optimization (GRPO) — 1× in this paper · explained in 6 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 8 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reward Function — 1× in this paper · explained in 4 episodes

Sources

• ClinicalBERT: Modeling Clinical Notes and Predicting Hospital Readmission
• MedGemma Technical Report
• Playing Atari with Deep Reinforcement Learning
• Reinforcement Learning through Asynchronous Advantage Actor-Critic on a GPU
• Proximal Policy Optimization Algorithms
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• From Reasoning to Code: GRPO Optimization for Underrepresented Languages
• RARL: Improving Medical VLM Reasoning and Generalization with Reinforcement Learning and LoRA under Data and Hardware Constraints
• Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
• HealthBench: Evaluating Large Language Models Towards Improved Human Health

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering