SOLVE-Med: Specialized Orchestration for Leading Vertical Experts across Medical Specialties

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Di Marino, Roberta, Dioguardi, Giovanni, Romano, Antonio, Riccio, Giuseppe, Barone, Mariano, Postiglione, Marco, Amato, Flora, Moscato, Vincenzo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909887919816704
author Di Marino, Roberta
Dioguardi, Giovanni
Romano, Antonio
Riccio, Giuseppe
Barone, Mariano
Postiglione, Marco
Amato, Flora
Moscato, Vincenzo
author_facet Di Marino, Roberta
Dioguardi, Giovanni
Romano, Antonio
Riccio, Giuseppe
Barone, Mariano
Postiglione, Marco
Amato, Flora
Moscato, Vincenzo
contents Medical question answering systems face deployment challenges including hallucinations, bias, computational demands, privacy concerns, and the need for specialized expertise across diverse domains. Here, we present SOLVE-Med, a multi-agent architecture combining domain-specialized small language models for complex medical queries. The system employs a Router Agent for dynamic specialist selection, ten specialized models (1B parameters each) fine-tuned on specific medical domains, and an Orchestrator Agent that synthesizes responses. Evaluated on Italian medical forum data across ten specialties, SOLVE-Med achieves superior performance with ROUGE-1 of 0.301 and BERTScore F1 of 0.697, outperforming standalone models up to 14B parameters while enabling local deployment. Our code is publicly available on GitHub: https://github.com/PRAISELab-PicusLab/SOLVE-Med.
format Preprint
id arxiv_https___arxiv_org_abs_2511_03542
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SOLVE-Med: Specialized Orchestration for Leading Vertical Experts across Medical Specialties
Di Marino, Roberta
Dioguardi, Giovanni
Romano, Antonio
Riccio, Giuseppe
Barone, Mariano
Postiglione, Marco
Amato, Flora
Moscato, Vincenzo
Computation and Language
Artificial Intelligence
Medical question answering systems face deployment challenges including hallucinations, bias, computational demands, privacy concerns, and the need for specialized expertise across diverse domains. Here, we present SOLVE-Med, a multi-agent architecture combining domain-specialized small language models for complex medical queries. The system employs a Router Agent for dynamic specialist selection, ten specialized models (1B parameters each) fine-tuned on specific medical domains, and an Orchestrator Agent that synthesizes responses. Evaluated on Italian medical forum data across ten specialties, SOLVE-Med achieves superior performance with ROUGE-1 of 0.301 and BERTScore F1 of 0.697, outperforming standalone models up to 14B parameters while enabling local deployment. Our code is publicly available on GitHub: https://github.com/PRAISELab-PicusLab/SOLVE-Med.
title SOLVE-Med: Specialized Orchestration for Leading Vertical Experts across Medical Specialties
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2511.03542