mlsanityData to 7 Oct 2026

medR: Reward Engineering for Clinical Offline Reinforcement Learning via Tri-Drive Potential Functions

arXiv 2602.03305 ↗First posted 3 Feb 2026cs.LGcited by 0

On the map: Offline RL and imitation learning from data → Safe offline RL

Abstract (arXiv)

Reinforcement Learning (RL) offers a powerful framework for optimizing dynamic treatment regimes (DTRs). However, clinical RL is fundamentally bottlenecked by reward engineering: the challenge of defining signals that safely and effectively guide policy learning in complex, sparse offline environments. Existing approaches often rely on manual heuristics that fail to generalize across diverse pathologies. To address this, we propose an automated pipeline leveraging Large Language Models (LLMs) for offline reward design and verification. We formulate the reward function using potential functions consisted of three core components: survival, confidence, and competence. We further introduce quantitative metrics to rigorously evaluate and select the optimal reward structure prior to deployment. By integrating LLM-driven domain knowledge, our framework automates the design of reward functions for specific diseases while significantly enhancing the performance of the resulting policies.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Safe offline RL, one of 13 directions in Offline RL and imitation learning from data. Builds on 32 of its 86 references that are in the map. First posted on arXiv 3 Feb 2026.

Builds on

32 of 86 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Advancing Safe Mechanical Ventilation Using Offline RL With Hybrid Actions and Clinically Aligned Rewards17 Jun

Other references · 31

  1. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 1,916
  2. 2021Offline Reinforcement Learning with Implicit Q-Learningcited by 1,245
  3. 2018Off-Policy Deep Reinforcement Learning without Explorationcited by 1,227
  4. 2023Eureka: Human-Level Reward Design via Coding Large Language Modelscited by 542
  5. 2023Reward Design with Language Modelscited by 257
  6. 2023Text2Reward: Reward Shaping with Language Models for Reinforcement Learningcited by 153
  7. 2021Reward (Mis)design for Autonomous Drivingcited by 102
  8. 2025Reward Shaping to Mitigate Reward Hacking in RLHFcited by 88
  9. 2024On Designing Effective RL Reward at Training Time for LLM Reasoningcited by 87
  10. 2017Continuous State-Space Models for Optimal Sepsis Treatment - a Deep Reinforcement Learning Approachcited by 68
  11. 2025SPA-RL: Reinforcing LLM Agents via Stepwise Progress Attributioncited by 68
  12. 2024The Ethics of ChatGPT in Medicine and Healthcare: A Systematic Review on Large Language Models (LLMs)cited by 45
19 more
  1. 2019Improving Sepsis Treatment Strategies by Combining Deep and Kernel-Based Reinforcement Learningcited by 34
  2. 2024Transformers Can Learn Temporal Difference Methods for In-Context Reinforcement Learningcited by 22
  3. 2024Latent Reward: LLM-Empowered Credit Assignment in Episodic Reinforcement Learningcited by 22
  4. 2018Model-Based Reinforcement Learning for Sepsis Treatmentcited by 21
  5. 2023Reinforcement Learning in the Era of LLMs: What is Essential? What is needed? An RL Perspective on RLHF, Prompting, and Beyondcited by 20
  6. 2025The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvementcited by 19
  7. 2017An Improved Multi-Output Gaussian Process RNN with Real-Time Validation for Early Sepsis Detectioncited by 16
  8. 2024The RL/LLM Taxonomy Tree: Reviewing Synergies Between Reinforcement Learning and Large Language Modelscited by 15
  9. 2024How Can LLM Guide RL? A Value-Based Approachcited by 13
  10. 2024Potential-Based Reward Shaping For Intrinsic Motivationcited by 12
  11. 2024G-Transformer: Counterfactual Outcome Prediction under Dynamic and Time-varying Treatment Regimescited by 8
  12. 2024Extracting Heuristics from Large Language Models for Reward Shaping in Reinforcement Learningcited by 7
  13. 2024Assessing the Zero-Shot Capabilities of LLMs for Action Evaluation in RLcited by 7
  14. 2025QLLM: Do We Really Need a Mixing Network for Credit Assignment in Multi-Agent Reinforcement Learning?cited by 6
  15. 2024Reward Guidance for Reinforcement Learning Tasks Based on Large Language Models: The LMGT Frameworkcited by 4
  16. 2023UGAE: A Novel Approach to Non-exponential Discountingcited by 3
  17. 2024From Reward Shaping to Q-Shaping: Achieving Unbiased Learning with LLM-Guided Knowledgecited by 3
  18. 2025Zero-Shot LLMs in Human-in-the-Loop RL: Replacing Human Feedback for Reward Shapingcited by 3
  19. 2025Are Large Language Models Dynamic Treatment Planners? An In Silico Study from a Prior Knowledge Injection Anglecited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026MediX-R1: Open Ended Medical Reinforcement Learning
  2. 2024Reinforcement Learning in Dynamic Treatment Regimes Needs Critical Reexamination
  3. 2026Learning to Compress Time-to-Control: A Reinforcement Learning Framework for Chronic Disease Management
  4. 2026Confounding Masquerading as Improvement: A Systematic Evaluation of Offline Reinforcement Learning for Stroke Antithrombotic Treatment in a 129,000-Patient Registry
  5. 2021Model Selection for Offline Reinforcement Learning: Practical Considerations for Healthcare Settings
  6. 2024RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMs
  7. 2025MedGR2: Breaking the Data Barrier for Medical Reasoning via Generative Reward Learning
  8. 2024OMG-RL:Offline Model-based Guided Reward Learning for Heparin Treatment