# Open (Clinical) LLMs are Sensitive to Instruction Phrasings

arXiv: https://arxiv.org/abs/2407.09429 · HTML: https://mlsanity.com/paper/2407.09429/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 53 references that are in the map.
- First posted on arXiv 12 Jul 2024.

## Builds on (most cited first, 29 of 29)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Multitask Prompted Training Enables Zero-Shot Task Generalization](https://mlsanity.com/paper/2110.08207/) (2021)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [ClinicalBERT: Modeling Clinical Notes and Predicting Hospital Readmission](https://mlsanity.com/paper/1904.05342/) (2019)
- [Do Prompt-Based Models Really Understand the Meaning of their Prompts?](https://mlsanity.com/paper/2109.01247/) (2021)
- [Multitask learning and benchmarking with clinical time series data](https://mlsanity.com/paper/1703.07771/) (2017)
- [MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data](https://mlsanity.com/paper/2304.08247/) (2023)
- [Adapted Large Language Models Can Outperform Medical Experts in Clinical Text Summarization](https://mlsanity.com/paper/2309.07430/) (2023)
- [OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalization](https://mlsanity.com/paper/2212.12017/) (2022)
- [Large Language Models are Few-Shot Clinical Information Extractors](https://mlsanity.com/paper/2205.12689/) (2022)
- [Do LLMs exhibit human-like response biases? A case study in survey design](https://mlsanity.com/paper/2311.04076/) (2023)
- [Do We Still Need Clinical Language Models?](https://mlsanity.com/paper/2302.08091/) (2023)
- [Thinking about GPT-3 In-Context Learning for Biomedical IE? Think Again](https://mlsanity.com/paper/2203.08410/) (2022)
- [Evaluating the Zero-shot Robustness of Instruction-tuned Language Models](https://mlsanity.com/paper/2306.11270/) (2023)
- [Publicly Shareable Clinical Large Language Model Built on Synthetic Clinical Notes](https://mlsanity.com/paper/2309.00237/) (2023)
- [Write It Like You See It: Detectable Differences in Clinical Notes By Race Lead To Differential Model Recommendations](https://mlsanity.com/paper/2205.03931/) (2022)
- [Zero-shot Approach to Overcome Perturbation Sensitivity of Prompts](https://mlsanity.com/paper/2305.15689/) (2023)
- [Jointly Extracting Interventions, Outcomes, and Findings from RCT Reports with LLMs](https://mlsanity.com/paper/2305.03642/) (2023)
- [On the Impact of Random Seeds on the Fairness of Clinical Classifiers](https://mlsanity.com/paper/2104.06338/) (2021)
- [ToNER: Type-oriented Named Entity Recognition with Generative Language Model](https://mlsanity.com/paper/2404.09145/) (2024)
- [Generalist embedding models are better at short-context clinical semantic search than specialized embedding models](https://mlsanity.com/paper/2401.01943/) (2024)

## Built on by (most cited first, 24 of 24)

- [Stereotype or Personalization? User Identity Biases Chatbot Recommendations](https://mlsanity.com/paper/2410.05613/) (2024)
- [Medical Adaptation of Large Language and Vision-Language Models: Are We Making Progress?](https://mlsanity.com/paper/2411.04118/) (2024)
- [A Comprehensive Survey on the Trustworthiness of Large Language Models in Healthcare](https://mlsanity.com/paper/2502.15871/) (2025)
- [ALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning](https://mlsanity.com/paper/2502.14860/) (2025)
- [Do Activation Verbalization Methods Convey Privileged Information?](https://mlsanity.com/paper/2509.13316/) (2025)
- [Elucidating Mechanisms of Demographic Bias in LLMs for Healthcare](https://mlsanity.com/paper/2502.13319/) (2025)
- [FactEHR: A Dataset for Evaluating Factuality in Clinical Notes Using LLMs](https://mlsanity.com/paper/2412.12422/) (2024)
- [Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models](https://mlsanity.com/paper/2506.14092/) (2025)
- [Retrieval-Augmented Agentic Rubric Generation for Reliable Medical Response Evaluation](https://mlsanity.com/paper/2601.15161/) (2026)
- [Explaining GPTs' Schema of Depression: A Machine Behavior Analysis](https://mlsanity.com/paper/2411.13800/) (2024)
- [Ensemble Debiasing Across Class and Sample Levels for Fairer Prompting Accuracy](https://mlsanity.com/paper/2503.05157/) (2025)
- [Evidence-Driven Decision Support for AI Model Selection in Research Software Engineering](https://mlsanity.com/paper/2512.11984/) (2025)
- [ThReadMed-QA: A Multi-Turn Medical Dialogue Benchmark from Real Patient Questions](https://mlsanity.com/paper/2603.11281/) (2026)
- [Rule vs.Consequence: Dissociable Internal Representations of Moral Reasoning in LLMs](https://mlsanity.com/paper/2610.16371/) (2026)
- [Reduction of Supervision for Biomedical Knowledge Discovery](https://mlsanity.com/paper/2504.09582/) (2025)
- [When the Gold Standard Isn't Necessarily Standard: Challenges of Evaluating the Translation of User-Generated Content](https://mlsanity.com/paper/2512.17738/) (2025)
- [A Systematic Evaluation of Large Language Models for PTSD Severity Estimation: The Role of Contextual Knowledge and Modeling Strategies](https://mlsanity.com/paper/2602.06015/) (2026)
- [Prompts Without Evidence: How Neuroimaging Mentions Shift Clinical Vision-Language Model Predictions](https://mlsanity.com/paper/2603.28387/) (2026)
- [Green Shielding: A User-Centric Approach Towards Trustworthy AI](https://mlsanity.com/paper/2604.24700/) (2026)
- [When Large Language Models Fail in Healthcare: Evaluating Sensitivity to Prompt Variations](https://mlsanity.com/paper/2606.07237/) (2026)
- [When Does Synthetic CT Transfer? A Label-Free Donor/Host Diagnostic for Medical Vision-Language Model Routing on Real Lung CT](https://mlsanity.com/paper/2606.29232/) (2026)
- [Evaluating Large Language Models on Misconceptions in Multi-Turn Medical Conversations](https://mlsanity.com/paper/2607.12884/) (2026)
- [Understanding Tone-Dependent Inference Cost in Large Language Models](https://mlsanity.com/paper/2607.23915/) (2026)
- [Epistemic Warrant for LLM Recommendations: Characterizing the Basis for Reliance When Ground Truth Is Unavailable](https://mlsanity.com/paper/2609.04127/) (2026)

## Abstract (arXiv)

Instruction-tuned Large Language Models (LLMs) can perform a wide range of tasks given natural language instructions to do so, but they are sensitive to how such instructions are phrased. This issue is especially concerning in healthcare, as clinicians are unlikely to be experienced prompt engineers and the potential consequences of inaccurate outputs are heightened in this domain. This raises a practical question: How robust are instruction-tuned LLMs to natural variations in the instructions provided for clinical NLP tasks? We collect prompts from medical doctors across a range of tasks and quantify the sensitivity of seven LLMs -- some general, others specialized -- to natural (i.e., non-adversarial) instruction phrasings. We find that performance varies substantially across all models, and that -- perhaps surprisingly -- domain-specific models explicitly trained on clinical data are especially brittle, compared to their general domain counterparts. Further, arbitrary phrasing differences can affect fairness, e.g., valid but distinct instructions for mortality prediction yield a range both in overall performance, and in terms of differences between demographic groups.