Research · The Decoder ·

Language models can't spark scientific revolutions, but world models might

Language models can't spark scientific revolutions, but world models might

In a position paper titled “LLMs Can’t Jump,” Google DeepMind researcher Tom Zahavy argues that language models lack a cognitive mechanism needed to generate fundamentally new scientific ideas. He suggests world models may be better suited to enabling scientific breakthroughs.

Read the full story at The Decoder →