AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
How Do You Come Up With a Benchmark for Analogy?
There's this linguistic based analogies, like word in bettings. And then there's the more idealized, non linguistic analogies, like or letter string examples. A lot of people now really are looking at what's missing from our current systems that we need to do this kind of abstract reasoning. It was part of a paper he wrote called on the measure of intelligence, which was about how do we compare ai systems with humans in a fair way?