# Prompt Sensitivity and Answer Consistency of Small Open-Source Language Models for Clinical Question Answering in Low-Resource Healthcare

arXiv: https://arxiv.org/abs/2603.00917 · HTML: https://mlsanity.com/paper/2603.00917/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 16 references that are in the map.
- First posted on arXiv 1 Mar 2026.

## Builds on (most cited first, 10 of 10)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [Medical Hallucinations in Foundation Models and Their Impact on Healthcare](https://mlsanity.com/paper/2503.05777/) (2025)
- [Towards LLMs Robustness to Changes in Prompt Format Styles](https://mlsanity.com/paper/2504.06969/) (2025)

## Abstract (arXiv)

Small open-source language models are gaining attention for healthcare applications in low-resource settings where cloud infrastructure and GPU hardware may be unavailable. However, the reliability of these models under different phrasings of the same clinical question remains poorly understood. We evaluate five open-source models (Gemma 2 2B, Phi-3 Mini 3.8B, Llama 3.2 3B, Mistral 7B, and Meditron-7B, a domain-pretrained model without instruction tuning) across three clinical question answering datasets (MedQA, MedMCQA, and PubMedQA) using five prompt styles: original, formal, simplified, roleplay, and direct. Model behavior is evaluated using consistency scores, accuracy, and instruction-following failure rates. All experiments were conducted locally on consumer CPU hardware without fine-tuning. Consistency and accuracy were largely independent across models. Gemma 2 achieved the highest consistency (0.845-0.888) but the lowest accuracy (33.0-43.5%), while Llama 3.2 showed moderate consistency (0.774-0.807) alongside the highest accuracy (49.0-65.0%). Roleplay prompts consistently reduced accuracy across all models, with Phi-3 Mini dropping 21.5 percentage points on MedQA. Meditron-7B exhibited near-complete instruction-following failure on PubMedQA (99.0% UNKNOWN rate), indicating that domain pretraining alone is insufficient for structured clinical question answering. These findings show that high consistency does not imply correctness: models can be reliably wrong, a dangerous failure mode in clinical AI. Llama 3.2 demonstrated the strongest balance of accuracy and reliability for low-resource deployment. Safe clinical AI requires joint evaluation of consistency, accuracy, and instruction adherence.