# LLM-Guided Evolution for Medical Decision Pipelines

arXiv: https://arxiv.org/abs/2606.07342 · HTML: https://mlsanity.com/paper/2606.07342/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 44 references that are in the map.
- First posted on arXiv 5 Jun 2026.

## Builds on (most cited first, 18 of 18)

- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [Illuminating search spaces by mapping elites](https://mlsanity.com/paper/1504.04909/) (2015)
- [MediQ: Question-Asking LLMs and a Benchmark for Reliable Interactive Clinical Reasoning](https://mlsanity.com/paper/2406.00922/) (2024)
- [Medical Image Understanding with Pretrained Vision Language Models: A Comprehensive Study](https://mlsanity.com/paper/2209.15517/) (2022)
- [Automated Machine Learning: From Principles to Practices](https://mlsanity.com/paper/1810.13306/) (2018)
- [ALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning](https://mlsanity.com/paper/2502.14860/) (2025)
- [A Survey of Sustainability in Large Language Models: Applications, Economics, and Challenges](https://mlsanity.com/paper/2412.04782/) (2024)
- [3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark](https://mlsanity.com/paper/2504.13861/) (2025)
- [GigaEvo: An Open Source Optimization Framework Powered By LLMs And Evolution Algorithms](https://mlsanity.com/paper/2511.17592/) (2025)
- [Large Language Models as In-context AI Generators for Quality-Diversity](https://mlsanity.com/paper/2404.15794/) (2024)
- [AutoMedPrompt: A New Framework for Optimizing LLM Medical Prompts Using Textual Gradients](https://mlsanity.com/paper/2502.15944/) (2025)
- [EMPOWER: Evolutionary Medical Prompt Optimization With Reinforcement Learning](https://mlsanity.com/paper/2508.17703/) (2025)
- [From Generalist to Specialist: Adapting Vision Language Models via Task-Specific Visual Instruction Tuning](https://mlsanity.com/paper/2410.06456/) (2024)
- [Enhancing the Medical Context-Awareness Ability of LLMs via Multifaceted Self-Refinement Learning](https://mlsanity.com/paper/2511.10067/) (2025)
- [From Pre-labeling to Production: Engineering Lessons from a Machine Learning Pipeline in the Public Sector](https://mlsanity.com/paper/2511.01545/) (2025)
- [PatientVLM Meets DocVLM: Pre-Consultation Dialogue Between Vision-Language Models for Efficient Diagnosis](https://mlsanity.com/paper/2601.10945/) (2026)

## Abstract (arXiv)

Adapting large language models (LLMs) to clinical workflows often requires costly fine-tuning or manual prompt and pipeline engineering. We study LLM-guided MAP-Elites evolution as an inference-time alternative for discovering medical decision strategies and provide an implementation repository at https://github.com/univanxx/llmguidedevomedical. We formulate urgency triage, interactive consultation, and medical image classification as evolutionary searches over executable artifacts optimized by task-specific fitness functions. Across all three settings, evolution improves over manually designed baselines under practical constraints. In triage, evolved programs increase Semigran accuracy from 77.3% to 87.1% and emergency recall from 0.60 to 0.97, while improving safety-weighted held-out MIMIC-ESI performance. In interactive consultation, evolved policies improve the accuracy--cost frontier across Llama-3, Qwen-3.5, and Gemma-4 and transfer to held-out iCRAFTMD. In PneumoniaMNIST, prompt-only evolution improves frozen MedGemma VLMs while preserving strict JSON outputs. Qualitative analysis shows that the gains come from interpretable program-level mechanisms, calibrated triage boundaries, targeted evidence acquisition, selective commitment, and finding-oriented visual decision rules, rather than superficial prompt rewording alone.