E245 Small Language Models, Contextual Intelligence and AI Sovereignty with Guglielmo Iozzia
Notas do episódio
Today's guest is Guglielmo Iozzia, Director, ML/AI and Applied Mathematics at MSD. In the ever-evolving domain of artificial intelligence, small language models pose a revolutionary alternative to the expansive and resource-intensive large-scale models. In this enlightening podcast episode, Guglielmo shares insights from his new book, "Domain-Specific Small Language Models: Efficient AI for Local Deployment,” as we dive deep into the advantages and challenges of small language models.
Topics include:
0:00 Optimising Small Language Models for Real-World Edge Applications
3:39 When Small Models Outperform Large Ones
6:16 Fine-Tuning AI with Domain-Specific Data
7:23 Why AI Benchmarks Don't Tell the Full Story
9:02 Lessons Learned After Writing the Book
10:52 Contextual Intelligence in Agentic AI Systems
12:23 How AI Books Are Written and Reviewed in PracticeLink to buy Guglielmo's latest book: https://www.amazon.com/Domain-Specific-Language-Models-Guglielmo-Iozzia/dp/1633436705