Corey Knowles

Cool.

Well, I have a question that's kind of a different direction that I've been really curious about, and that's, does diffusion change hallucination behavior or controllability of a model, or is it similar in nature?

Interpretability has become such a fascinating field in terms of just looking at, you know, yeah, we know these are the parts.

We know this is how we build it.

We know that these principles work on this end and these principles work on this end.

But somewhere in the middle, is that like,

like leap of faith that that we don't quite get uh and uh it's really it's fascinating grant i'm glad you asked that question so much you gotta you just got you just gotta ask you know it's like that's right i don't know so gotta ask the people who know so do you feel like

Does model evaluation, is it approached any differently with diffusion or are we essentially just looking at the answers in there?

I just, I wasn't sure.

I wrote this question down and I wasn't sure it wasn't really stupid, but I decided I was going to ask it anyways, as if not being sequential effects, how those metrics work and evals.

You have to track them, but also respect that benchmarks need a bit of a grain of salt with them as well.

In the end, it's down to your tasks and what you're doing where you find that, I guess.

Well, I know we're getting tight on time, but I have one last question before I let you go.

And I want to I want to make sure we talk a little bit more about Mercury because it's been around a little minute, a minute now.

And it's surprisingly good for something that some people don't even know exists at this point.

And I'm wondering, like.

What's what's on the roadmap?

What's.

What's coming up in the near future?

Maybe the longer term future?