Development of Continual Routing Methodologies for Foundation Models
Scientific-Disciplinary Group
01/INFO-01 - Informatics
Description
Artificial Intelligence, powered by foundational models, enables highly adaptive systems, yet optimal selection and integration remain complex, inefficient, and poorly interoperable. This project develops Continuous Routing methodologies for the automatic selection of the most suitable foundational AI models for each context, optimizing performance, computational cost, and scalability. The proposed framework is modular and adaptive, enabling dynamic routing among models based on performance and context. The project involves writing one or more scientific articles that describe and empirically evaluate a methodology that reflects these requirements.
Compensation
28,456 Euro
Job posting website
Number of positions
1
Maximum duration
12.0
Funding body
Luiss Guido Carli
View the original posting on the MUR website: Go to MUR website