Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Aithal, Manjushree B., D., Ph., Kotz, Alexander, Mitchell, James, D, Ph.
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.23678
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914420276330496
author Aithal, Manjushree B.
D., Ph.
Kotz, Alexander
Mitchell, James
D, Ph.
author_facet Aithal, Manjushree B.
D., Ph.
Kotz, Alexander
Mitchell, James
D, Ph.
contents Large Language Models (LLMs) offer transformative solutions across many domains, but healthcare integration is hindered by strict data privacy constraints. Clinical narratives are dense with ambiguous acronyms, misinterpretation these abbreviations can precipitate severe outcomes like life-threatening medication errors. While cloud-dependent LLMs excel at Acronym Disambiguation, transmitting Protected Health Information to external servers violates privacy frameworks. To bridge this gap, this study pioneers the evaluation of small-parameter models deployed entirely on-device to ensure privacy preservation. We introduce a privacy-preserving cascaded pipeline leveraging general-purpose local models to detect clinical acronyms, routing them to domain-specific biomedical models for context-relevant expansions. Results reveal that while general instruction-following models achieve high detection accuracy (~0.988), their expansion capabilities plummet (~0.655). Our cascaded approach utilizes domain-specific medical models to increase expansion accuracy to (~0.81). This novel work demonstrates that privacy-preserving, on-device (2B-10B) models deliver high-fidelity clinical acronym disambiguation support.
format Preprint
id arxiv_https___arxiv_org_abs_2603_23678
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PLACID: Privacy-preserving Large language models for Acronym Clinical Inference and Disambiguation
Aithal, Manjushree B.
D., Ph.
Kotz, Alexander
Mitchell, James
D, Ph.
Computation and Language
Artificial Intelligence
Large Language Models (LLMs) offer transformative solutions across many domains, but healthcare integration is hindered by strict data privacy constraints. Clinical narratives are dense with ambiguous acronyms, misinterpretation these abbreviations can precipitate severe outcomes like life-threatening medication errors. While cloud-dependent LLMs excel at Acronym Disambiguation, transmitting Protected Health Information to external servers violates privacy frameworks. To bridge this gap, this study pioneers the evaluation of small-parameter models deployed entirely on-device to ensure privacy preservation. We introduce a privacy-preserving cascaded pipeline leveraging general-purpose local models to detect clinical acronyms, routing them to domain-specific biomedical models for context-relevant expansions. Results reveal that while general instruction-following models achieve high detection accuracy (~0.988), their expansion capabilities plummet (~0.655). Our cascaded approach utilizes domain-specific medical models to increase expansion accuracy to (~0.81). This novel work demonstrates that privacy-preserving, on-device (2B-10B) models deliver high-fidelity clinical acronym disambiguation support.
title PLACID: Privacy-preserving Large language models for Acronym Clinical Inference and Disambiguation
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2603.23678