2026-06-02原文2 min readUpdated: 2026-06-02

A Multi-Domain Red Teaming Framework for Safety, Robustness, and Fairness Evaluation of Medical Large Language Models

arXiv:2606.00027v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed across healthcare, yet existing benchmarks fail to capture model behavior under adversarial or ethically complex conditions common in clinical practice. We developed a multi-domain red teaming framework evaluating eleven contemporary LLMs across 690 clinically grounded scenarios spanning nine domains and over 150 subcategories. Scenarios incorporated adversarial transformations, and responses were assessed using a seven-dimension rubric with LLM-assisted scoring and human-in-the-loop validation. Results revealed substantial performance variance, with mean scores ranging from 0.791 to 0.984. Critically, several high-performing systems produced complete failures in individual safety-critical scenarios, demonstrating that aggregate accuracy masks clinically meaningful risk. The highest-performing systems (X-BAI, GPT-5, Claude Opus 4.1) achieved scores above 0.97 with low variance, while performance varied significantly across domains. Equity-related tasks showed 10-20% error amplification with demographic modifications, and human reviewers identified clinically relevant failures missed by automated evaluation. Our findings demonstrate that performance variance and worst-case failures provide more clinically meaningful reliability indicators than mean accuracy alone, and that hybrid evaluation approaches combining automation with clinician oversight are essential for credible safety assessment.

SourcearXiv Computational LinguisticsAuthor: Andrei Marian Feier, Veysel Kocaman, Yigit Gul, Ahmet Korkmaz, Alexander Thomas, Aleksei Zakharov, Jay Gil, Mehmet Butgul, David Talby

[2606.00027] A Multi-Domain Red Teaming Framework for Safety, Robustness, and Fairness Evaluation of Medical Large Language Models

[Submitted on 15 Apr 2026]

Title:A Multi-Domain Red Teaming Framework for Safety, Robustness, and Fairness Evaluation of Medical Large Language Models

View a PDF of the paper titled A Multi-Domain Red Teaming Framework for Safety, Robustness, and Fairness Evaluation of Medical Large Language Models, by Andrei Marian Feier and 8 other authors

View PDF HTML (experimental)

Abstract:Large language models (LLMs) are increasingly deployed across healthcare, yet existing benchmarks fail to capture model behavior under adversarial or ethically complex conditions common in clinical practice. We developed a multi-domain red teaming framework evaluating eleven contemporary LLMs across 690 clinically grounded scenarios spanning nine domains and over 150 subcategories. Scenarios incorporated adversarial transformations, and responses were assessed using a seven-dimension rubric with LLM-assisted scoring and human-in-the-loop validation. Results revealed substantial performance variance, with mean scores ranging from 0.791 to 0.984. Critically, several high-performing systems produced complete failures in individual safety-critical scenarios, demonstrating that aggregate accuracy masks clinically meaningful risk. The highest-performing systems (X-BAI, GPT-5, Claude Opus 4.1) achieved scores above 0.97 with low variance, while performance varied significantly across domains. Equity-related tasks showed 10-20% error amplification with demographic modifications, and human reviewers identified clinically relevant failures missed by automated evaluation. Our findings demonstrate that performance variance and worst-case failures provide more clinically meaningful reliability indicators than mean accuracy alone, and that hybrid evaluation approaches combining automation with clinician oversight are essential for credible safety assessment.

Comments: 10 pages, 4 figures. To be presented at the Text2Story 2026 Workshop (Delft, The Netherlands, 29 March 2026); CEUR Workshop Proceedings (forthcoming). Affiliation: John Snow Labs Inc

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.00027 [cs.CL]

(or arXiv:2606.00027v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2606.00027

arXiv-issued DOI via DataCite

Submission history

From: Yigit Gul [view email] [v1] Wed, 15 Apr 2026 14:57:55 UTC (3,394 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled A Multi-Domain Red Teaming Framework for Safety, Robustness, and Fairness Evaluation of Medical Large Language Models, by Andrei Marian Feier and 8 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-06

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)