Preprint
Red Teaming Large Language Models for Healthcare
ArXiV.org
Cornell University
05/01/2025
DOI: 10.48550/arxiv.2505.00467
Abstract
We present the design process and findings of the pre-conference workshop at the Machine Learning for Healthcare Conference (2024) entitled Red Teaming Large Language Models for Healthcare, which took place on August 15, 2024. Conference participants, comprising a mix of computational and clinical expertise, attempted to discover vulnerabilities -- realistic clinical prompts for which a large language model (LLM) outputs a response that could cause clinical harm. Red-teaming with clinicians enables the identification of LLM vulnerabilities that may not be recognised by LLM developers lacking clinical expertise. We report the vulnerabilities found, categorise them, and present the results of a replication study assessing the vulnerabilities across all LLMs provided.
Details
- Title: Subtitle
- Red Teaming Large Language Models for Healthcare
- Creators
- Vahid Balazadeh - University of TorontoMichael Cooper - University of TorontoDavid Pellow - University of TorontoAtousa Assadi - University of TorontoJennifer BellMark Coastworth - Vector InstituteKaivalya Deshpande - NYU Langone HealthJim Fackler - Johns Hopkins UniversityGabriel Funingana - Cancer Research UK Cambridge InstituteSpencer Gable-CookAnirudh GangadharAbhishek JaiswalSumanth KajaChristopher KhouryAmrit Krishnan - Vector InstituteRandy Lin - Algoma UniversityKaden McKeen - University of TorontoSara Naimimohasses - University of IowaKhashayar Namdar - University of TorontoAviraj Newatia - University of TorontoAllan Pang - Leeds Teaching Hospitals NHS TrustAnshul PattooSameer PeesapatiDiana Prepelita - University of CambridgeBogdana RakovaSaba Sadatamin - University of TorontoRafael SchulmanAjay Shah - University of EdinburghSyed Azhar ShahSyed Ahmar ShahBabak Taati - University of TorontoBalagopal Unnikrishnan - University of TorontoStephanie WilliamsRahul G Krishnan - University of Toronto
- Resource Type
- Preprint
- Publication Details
- ArXiV.org
- DOI
- 10.48550/arxiv.2505.00467
- ISSN
- 2331-8422
- Publisher
- Cornell University; Ithaca, New York
- Language
- English
- Date posted
- 05/01/2025
- Academic Unit
- Gastroenterology and Hepatology; Internal Medicine
- Record Identifier
- 9984816017402771
Metrics
14 Record Views