How to Expand the Capabilities of Large Language Models

Watch a 38-minute tech talk by Jonathan Mugan, Principal Scientist at DeUmbra, exploring how to enhance the capabilities of Large Language Models (LLMs). Dive deep into the internal workings of LLMs and discover methods for expanding their world understanding through cognitive foundation training. Learn about combining LLMs with Good Old-Fashioned Artificial Intelligence (GOFAI) to leverage the strengths of both approaches - the flexibility of LLMs and the precision of GOFAI. Understand how training LLMs to generate GOFAI representations like logic formulas can create more robust AI systems that overcome current limitations in truth, exactness, and novel thinking.