mlsanityData to 14 Oct 2026

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

arXiv 2606.05174 ↗First posted 17 Apr 2026cs.CL cs.AIcited by 0

On the map: LLM reasoning → Clinical and biomedical applications

Abstract (arXiv)

Large language models can support medical question answering, but privacy, cost, and reliability still limit their clinical use in focused domains. Can rubric-based reinforcement learning bring a locally deployable model close to much larger systems? We study Group Relative Policy Optimization (GRPO) for heart-focused medical question answering with rubric supervision from RaR-Medicine. Rather than rely on binary rubric aggregation or a single overall score, our variance-aware reward framework builds continuous rewards from criterion-level outcomes. It preserves partial credit and reduces reward sparsity. We adapt a Qwen3-14B policy with low-rank adapters and quantized weights. On a held-out heart-related HealthBench subset, the best GRPO variant improves accuracy from 0.362 to 0.502 and F1 from 0.532 to 0.668 over the base model, close to GPT-OSS-120B. An independent clinician rated most responses clinically acceptable (61%) and patient ready (63%), both above the automated accuracy; fewer than 2% were unsafe. Taken together, the results support rubric-based reinforcement learning as a practical route for focused medical language models.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning. Builds on 20 of its 45 references that are in the map. First posted on arXiv 17 Apr 2026.

Builds on

20 of 45 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025RARL: Improving Medical VLM Reasoning and Generalization with Reinforcement Learning and LoRA under Data and Hardware Constraints7 Jun
  2. 2025MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning26 Feb

Other references · 18

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  6. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  7. 2015Trust Region Policy Optimizationcited by 2,580
  8. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,745
  9. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,184
  10. 2019ClinicalBERT: Modeling Clinical Notes and Predicting Hospital Readmissioncited by 493
  11. 2025MedGemma Technical Reportcited by 436
  12. 2025HealthBench: Evaluating Large Language Models Towards Improved Human Healthcited by 429
6 more
  1. 2025Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domainscited by 366
  2. 2016Reinforcement Learning through Asynchronous Advantage Actor-Critic on a GPUcited by 48
  3. 2025From Reasoning to Code: GRPO Optimization for Underrepresented Languagescited by 11
  4. 2025DeepTrans: Deep Reasoning Translation via Reinforcement Learningcited by 8
  5. 2024Personalizing Prostate Cancer Education for Patients Using an EHR-Integrated LLM Agentcited by 7
  6. 2021Modified Double DQN: addressing stabilitycited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026ConRub-Med: Reinforcement Learning with Consensus Rubrics for Open-Ended Medical Question Answering
  2. 2025Med-U1: Incentivizing Unified Medical Reasoning in LLMs via Large-scale Reinforcement Learning
  3. 2026Fathom-Vaidya: Advancing Medical Reasoning with Rubric-Based Rewards
  4. 2026Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks
  5. 2025Baichuan-M2: Scaling Medical Capability with Large Verifier System
  6. 2025Toward Effective Reinforcement Learning Fine-Tuning for Medical VQA in Vision-Language Models
  7. 2026Gaokerena: A Small Persian Medical Language Model Family
  8. 2025RPRO: Ranked Preference Reinforcement Optimization for Enhancing Medical QA and Diagnostic Reasoning