Large Language Models’ Emergent Abilities Are a Mirage

The original version of this story appeared in Quanta Magazine.

Two years ago, in a project called the Beyond the Imitation Game benchmark, or BIG-bench, 450 researchers compiled a list of 204 tasks designed to test the capabilities of large language models, which power chatbots like ChatGPT. On most tasks, performance improved predictably and smoothly as the models scaled up—the larger the model, the better it got. But with other tasks, the jump in ability wasn’t smooth. The performance remained near zero for a while, then performance jumped. Other studies found similar leaps in ability.

The authors described this as “breakthrough” behavior; other researchers have likened

→ Continue reading at Wired - Science

More from author

Related posts

Advertisment

Latest posts

Goldman Sachs analysts say there’s still potential for more stock gains | CNN Business

New York CNN  —  It’s been a bumpy road for the stock market this year, but investors aren’t complaining....

Chuck Todd rips NBC News for hiring former RNC chair Ronna McDaniel | CNN Business

CNN  —  Former “Meet the Press” moderator Chuck Todd delivered a stunning rebuke of NBC News on Sunday,...

Learn ASL with This Bundle, Just $40 Through March 24

Disclosure: Our goal is to feature products and services that we think you'll find interesting and useful. If you purchase them, Entrepreneur may...