3 ms·

This really misrepresents what is happening in these models. Complex semantics are represented in the latent space. Relationships between embeddings have meanin

by causal 3y ago

This really misrepresents what is happening in these models. Complex semantics are represented in the latent space. Relationships between embeddings have meanings we're able to sample to get these probabilistic strings - the fact that probabilities are involved does not mean there is no understanding behind the output.

mannykannot 3y ago
> Complex semantics are represented in the latent space. In this particular case, what's going on doesn't seem all that complex. > The fact that probabilities are involved does not mean there is no understanding behind the output. And that, in turn, does not mean that there is - which is the question of interest here. The burden of proof is on anyone claiming to know the answer, either way.