In this series of educational videos aimed at the general public, the Flowers research team from the Inria Center at the University of Bordeaux , specialized in Artificial Intelligence (AI), invites you to better understand how language models like ChatGPT work.
Recognizing that the vast majority of educational resources on these models adopted a lengthy format aimed at a relatively knowledgeable audience, this project of short-format videos is particularly geared towards middle and high school students, as well as educators, and more generally towards non-specialists in computer science or AI.

Episode 1

thumbnail

This video is an introduction to large language models which are at the core of tools like ChatGPT or Bard.

*
Download:

Episodes


Episode 2

thumbnail

Language Models, such as ChatGPT or Bard, have shown impressive capabilities. However, they also have numerous limitations and weaknesses!

*
Download:

Episode 3

thumbnail

Prompting, or how to make a language model perform a given task.

*
Download:

*These videos are royalty-free, with a Creative Commons CC-BY license. This means that they can be used freely, provided that their source is cited (see Credits). The voice in the english version of the series was synthesized using ElevenLabs (elevenlabs.io).

About


The Flowers team at the Inria Center of the University of Bordeaux conducts scientific research in the field of developmental Artificial Intelligence. In particular, they are interested in the computer modeling of human learning mechanisms and the roles of curiosity and language in cognitive development. They also study the properties of large language models (ChatGPT, Mistral, etc.), and develop methods to better anchor them in the real world.

The team also investigates applications of these works in the field of educational technologies, as well as to assist scientists (physicists, biologists, etc.) in exploring and better understanding new systems (materials, self-organized chemical reactions, etc.).