-
achieved: The work plan will consist of: i) improving the development of computational tools and optimizing the parameters of the best tool for detecting fall risk situations in assisted walking with
-
(LLMs); Configure and optimize cloud computing solutions or on-premise infrastructures that ensure high availability and scalability; Implement tools for efficient resource management, such as GPU
-
“Intelligent Models for Outpatient and Medical Exams Scheduling Optimization”, reference FCT 2024.07481.IACDC/2024, financed by measure RE-C05-i08. M04 – "Support the launch of an R&D project programme aimed
-
Preparation of the Technological Infrastructure:Design, configure and optimize the hardware and software infrastructure necessary for the training, fine-tuning and deployment of LLMs on a large scale; Explore
-
. The scholarship work plan will also include support for determining the specifications of a textile interface where the actuator will be integrated, namely the distribution of stiffness necessary to optimize force
-
to be developed consists of: Data Structuring and Optimization for Language Models:Implement routines for the collection, cleaning, and pre-processing of educational data; Establish methodologies
-
, and teaching approaches, optimizing the learning experience. Implementation of Motivational Elements and Immersive StorytellingDesign and apply gamification elements, such as point systems, achievements
-
datasets to train and fine-tune each adapter, optimizing it for the persona and content of that specific school level; Define metrics to evaluate the effectiveness of adapted communication. This will include
-
for the attribution of 2 grant of Master within the scope of the project of R&D “Intelligent Models for Outpatient and Medical Exams Scheduling Optimization”, FCT 2024.07481.IACDC/2024, supported by
-
the use of specialized models (e.g., for summarization, translation, question answering) in conjunction with core LLMs. Design of User-Centered Dialogues and Interactions:Design and optimize advanced prompt