Neuroimaging and Machine Learning for Enhancing Functional Recovery

​

Collaborating Institutions:

CENTRE HOSPITALIER UNIVERSITAIRE DE REIMS (CHU de Reims),

CLEMSON UNIVERSITY

METACOGNITION INSTITUTE

 

This research outlines a collaborative effort between the above stated research institutions to investigate the potential of advanced neuroimaging techniques and machine learning algorithms in understanding and enhancing recovery processes in individuals with functional disabilities.

 

Specifically, the project explores the analysis of brain signals and biomechanics. By combining the medical expertise and resources of CHU de Reims with the computational and analytical capabilities of Clemson University and the Metacognition Institute biometric and biofeedback research, we aim to develop innovative interventions that can improve the lives of those facing neurological challenges.

​

​

Understanding the limits of LLMs through an in depth analysis of human language. 

​

Collaborating Institutions:
CLEMSON UNIVERSITY

METACOGNITION INSTITUTE

​

We are designing a set of tests to understand the logical and reasoning limits of current LLMs. By carefully analysing the roots of language and its grounding, we aim to provide a framework to help us understand the role of language in the human scene.

​

Our approach involves dissecting the foundational aspects of language, including syntax, semantics, and pragmatics, to pinpoint where LLMs excel and where they fall short. By creating a comprehensive suite of benchmarks that reflect the complexity and nuances of human communication, we can better assess the capabilities and limitations of these models. This analysis will not only highlight the gaps in current LLMs but also guide future research and development towards more robust and human-like language processing systems.

​

Furthermore, we will explore how language is intertwined with human cognition and social interaction. Understanding the contextual and cultural elements that influence language use can reveal how LLMs interpret and generate text. This insight is crucial for developing LLMs that can navigate the intricacies of human language with greater accuracy and empathy. Our ultimate goal is to bridge the gap between artificial and human intelligence, fostering more meaningful and effective communication between humans and machines.

​