Edward Gibson: uncertainty

Transcript context

…Got it. And then what you’re saying is there’s a bunch of different kinds of experiments you can do. I mean, I’d like the understanding one is like have more embedding, more center embedding, is it easier or harder to understand? But then you have to measure the level of understanding, I guess? Yeah, you could. I mean there’s multiple ways to do that. I mean there’s the simplest way is just ask people how good does it sound? How natural does it sound? That’s a very blunt, but very good measure. It’s very reliable. People will do the same thing. And so it’s like, “I don’t know what it means exactly, but it’s doing something. ” Such that we’re measuring something about the confusion, the difficulty associated with those. And those are giving you a signal. That’s why you can say they’re… Okay. What about the completion with the center embedding?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.