Sussex Data Science is back!
Join us for our March Meetup featuring Colin Hayhurst & Lorenzo Bertolini.
Colin asks 💭 “Will Chatbots Kill the Search Star?” 🤔
Everyone, it seems, is playing with chatbots, and search engines are looking to get in on the action. But what lies underneath the AIs of chatbots? And why do they suddenly seem so adept at churning out plausible answers? In the talk, I’ll review the main LLMs (Large Language Models) that underlie chabots, and the innovations in ChatGPT. The main data sources behind these LLMs are sometimes transparent and often opaque. As a search engine and through our involvement in GAFAM chatbots since late 2021, we know more many about what lies behind the veil of secrecy. ChatGPT is no Google killer as some have suggested, but the race for AI supremacy poses some obvious, and also more subtle, threats to the web, as I shall explain.
🎤 Lorenzo presents : Transformers and Hugging Faces: theory and applications of NLP’s most notorious architecture
While the world is now very familiar with the name and concept of large language models, their (primary) theoretical and technical backbone, the transformer architecture, has received considerably fewer media attention. The transformer’s takeover in NLP research and practice has grown parallel with a particular resource, Hugging Face, which currently enables and supports thousands of free and open-source research and industry projects. In this talk, I will present an overview of transformers’ theoretical and practical foundations and how to integrate new and existing LLMs via Hugging Face, to create shareable and fully open-source NLP projects.