Communications

The rule of writing explained by - Harvard Professor - Steven Pinker 

"LRM" can refer to several different things depending on the context:

1. Large Reasoning Models (AI) 

Large Reasoning Models (LRMs) are an advanced type of Artificial Intelligence. 

They go beyond traditional Large Language Models (LLMs) by using reasoning. 

How they work: 

LRMs use "chain of thought" prompting to create a plan, evaluate options, and verify calculations. 

LLMs predict the next token based on statistics.

Purpose: LRMs are designed for complex tasks. These include math, coding, and logical reasoning. 

Examples: Deep-Seek R1, Open AI's o1/o3 series, and Gemini 2.0 Flash Thinking.

No comments:

Post a Comment