SOLVE-Med: Specialized Orchestration for Leading Vertical Experts across Medical Specialties
This addresses deployment challenges in medical AI systems, such as hallucinations and computational demands, by enabling efficient, specialized expertise across domains, though it is incremental as it builds on existing multi-agent and fine-tuning approaches.
The paper tackles the problem of medical question answering by proposing SOLVE-Med, a multi-agent architecture that combines specialized small language models, achieving superior performance with ROUGE-1 of 0.301 and BERTScore F1 of 0.697 on Italian medical forum data.
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.