CCoE: Approach to Mastering Multiple Domains with LLMs

Author(s): Manpreet Singh

Originally published on Towards AI.

This member-only story is on us. Upgrade to access all of Medium.

AI is changing quickly, and big language models (LLMs) like GPT-4 are making waves because they understand and create text very well.

But even though these models are great at general tasks, they often struggle with specific fields like medical advice, law, or advanced math.

A new framework called CCoE (Collaboration of Experts) is here to help fix that problem.

Let’s look at what CCoE is and why it could be important for the AI community.

Most of today’s LLMs are trained on lots of different kinds of data that makes them great generalists.

But for specific fields, like coding, hard math problems or giving accurate medical advice they can fall short. They don’t have enough deep, specialized knowledge.

Trying to make them better for each specific field takes a lot of computing power and time. Plus, it can cause a problem called catastrophic forgetting. Where the model forgets things it used to know after learning new things in a specific area.

This is where CCoE helps.

It lets us add specialized experts to a bigger LLM without losing the general knowledge and without huge computing costs.

Think of CCoE as a team working together.

Imagine a big,… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI