Get the latest tech news

Coconut by Meta AI – Better LLM Reasoning with Chain of Continuous Thought?


Discover how Meta AI’s Chain of Continuous Thought (Coconut) empowers large language models (LLMs) to reason in their own language

Neuroimaging studies have shown that the region of the human brain responsible for language comprehension and production remains largely inactive during reasoning tasks. Chain of Continuous Thought Multi-Stage Training Procedure ( Source)The researchers leverage existing language Chain-of-Thought data, where each sample consists of a question, reasoning steps, and the final answer. A possible explanation for this is that the thought tokens allow the model to explore multiple possible branches before committing to a specific path, whereas chain-of-thought reasoning chooses a direction from the start.

Get the Android app

Or read this on Hacker News

Read more on:

Photo of Meta AI

Meta AI

Photo of reasoning

reasoning

Photo of chain

chain

Related news:

News photo

Being a Developer in the Age of Reasoning AI

News photo

Google Releases Its Own 'Reasoning' AI Model

News photo

Google reveals AI ‘reasoning’ model that ‘explicitly shows its thoughts’