ReWoo: How to Decouple Reasoning of Language Models

Author(s): Reza Yazdanfar

Originally published on Towards AI.

Read this short, simple section, you'll get if this article suits you or not!! ????

Photo by Digital Content Writers India on Unsplash

ALM = ability of reasoning + reinforcement tools

ALM: Augmented Language Models

These tools help language models retrieve information and execute actions.

OK, cool, what’s the current approach?

Generating and then by tons of iteration trying to improve it.

OK OK, cool again, how’s the current approach?

Well, the current approach has a straightforward manner, easy to implement but computationally complex.

OK OK OK, cool … Any suggestion?

Yeah, actually, for the first time, an attempt has been made to combat.

Whats that??

ReWOO (Reasoning WithOut Observation)

Howz that??

Well, basically, it tries to split the reasoning process from external observation.

Consequently, it reduces token utilization.

The… 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