# Teaching LLMs Brazilian Healthcare: Injecting Knowledge from Official Clinical Guidelines

arXiv: https://arxiv.org/abs/2605.01077 · HTML: https://mlsanity.com/paper/2605.01077/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 35 references that are in the map.
- First posted on arXiv 1 May 2026.

## Builds on (most cited first, 25 of 25)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [Capabilities of GPT-4 on Medical Challenge Problems](https://mlsanity.com/paper/2303.13375/) (2023)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [LoRA Learns Less and Forgets Less](https://mlsanity.com/paper/2405.09673/) (2024)
- [BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains](https://mlsanity.com/paper/2402.10373/) (2024)
- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [Capabilities of Gemini Models in Medicine](https://mlsanity.com/paper/2404.18416/) (2024)
- [Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling](https://mlsanity.com/paper/2401.16380/) (2024)
- [Nemotron-CC: Transforming Common Crawl into a Refined Long-Horizon Pretraining Dataset](https://mlsanity.com/paper/2412.02595/) (2024)
- [LoRA vs Full Fine-tuning: An Illusion of Equivalence](https://mlsanity.com/paper/2410.21228/) (2024)
- [AlpaCare:Instruction-tuned Large Language Models for Medical Application](https://mlsanity.com/paper/2310.14558/) (2023)
- [Spurious Forgetting in Continual Learning of Language Models](https://mlsanity.com/paper/2501.13453/) (2025)
- [Adapting Large Language Models to Domains via Reading Comprehension](https://mlsanity.com/paper/2309.09530/) (2023)
- [Scaling Laws for Forgetting When Fine-Tuning Large Language Models](https://mlsanity.com/paper/2401.05605/) (2024)
- [Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection](https://mlsanity.com/paper/2502.06042/) (2025)
- [BLUEX: A benchmark based on Brazilian Leading Universities Entrance eXams](https://mlsanity.com/paper/2307.05410/) (2023)
- [Evaluating Parameter Efficient Methods for RLVR](https://mlsanity.com/paper/2512.23165/) (2025)
- [Adapting LLMs for the Medical Domain in Portuguese: A Study on Fine-Tuning and Model Evaluation](https://mlsanity.com/paper/2410.00163/) (2024)
- [HealthQA-BR: A System-Wide Benchmark Reveals Critical Knowledge Gaps in Large Language Models](https://mlsanity.com/paper/2506.21578/) (2025)
- [MedPT: A Massive Medical Question Answering Dataset for Brazilian-Portuguese Speakers](https://mlsanity.com/paper/2511.11878/) (2025)

## Abstract (arXiv)

Brazil's Unified Health System (SUS) relies on official clinical guidelines that define diagnostic criteria, treatments, dosages, and monitoring procedures for over 200 million citizens. Yet current LLMs perform poorly on this guideline-specific knowledge, and no benchmark evaluates clinical recall grounded in Brazilian Portuguese protocols. We address this gap by adapting Qwen2.5-14B-Instruct to the Brazilian clinical domain. From 178 official guidelines (~5.4M tokens), we generate ~70M tokens of synthetic data in three formats -- rephrases, wiki-style articles, and question-answer pairs -- using four generator LLMs. We then apply continual pre-training followed by Group Relative Policy Optimization (GRPO). We introduce HealthBench-BR, with 1,780 balanced true/false clinical assertions, and PCDT-QA, with 890 open-ended clinical questions scored by an LLM judge. Our best model achieves 83.9% on HealthBench-BR and 85.4% on PCDT-QA, outperforming GPT-5.2, Claude Sonnet 4.6, Gemini 3.1 Pro, and Google AI Overview's web-grounded RAG despite having only 14B parameters. Ablations show that generator diversity and reinforcement learning are critical to these gains. We release all datasets, benchmarks, and model weights to support reproducible clinical NLP research for Brazilian Portuguese. Code, data, and model weights are available at https://github.com/hugoabonizio/clinical-protocols-br