Get the latest tech news

LLMs don't do formal reasoning – and that is a HUGE problem


Important new study from Apple

Another manifestation of the lack of sufficiently abstract, formal reasoning in LLMs is the way in which performance often fall apart as problems are made bigger. Elon Musk’s putative robotaxis are likely to suffer from a similar affliction: they may well work sfaely for the most common situations, but are also likely struggle to reason abstractly enough in some circumstances. Gary Marcus is the author of The Algebraic Mind, a 2001 MIT Press Book that foresaw the Achilles’ Heel of current models.

Get the Android app

Or read this on Hacker News

Read more on:

Photo of LLMs

LLMs

Photo of huge problem

huge problem

Photo of HUGE

HUGE

Related news:

News photo

Health insurtech startup Qantev raises €30 million to outperform LLMs with small AI models

News photo

Pythagora: Auto-Generate Node.js Tests Using LLMs, No Coding Required

News photo

LLMs, Theory of Mind, and Cheryl's Birthday