A locally deployed Qwen3 4B model achieved 82.25% mean accuracy retrieving 56 features from pemphigus longitudinal records and generated summaries rated 7.93-8.50 by dermatologists with 53.3% preference over human versions.
and Yuan, William and Poddar, Mukund and Elsamadisi, Pansy and Brat, Gabriel A
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A narrative review of AI language technologies in multilingual healthcare identifies performance gaps in safety and equity and proposes seven grand challenges centered on reliability, human oversight, and cross-disciplinary collaboration.
citing papers explorer
-
Privacy-Preserving Local Language Models for Longitudinal Data Retrieval in Chronic Dermatologic Disease: Implementation in Pemphigus Patients
A locally deployed Qwen3 4B model achieved 82.25% mean accuracy retrieving 56 features from pemphigus longitudinal records and generated summaries rated 7.93-8.50 by dermatologists with 53.3% preference over human versions.
-
Artificial intelligence language technologies in multilingual healthcare: Grand challenges ahead
A narrative review of AI language technologies in multilingual healthcare identifies performance gaps in safety and equity and proposes seven grand challenges centered on reliability, human oversight, and cross-disciplinary collaboration.