Ethan He

speaker

722 appearances 1 recordings 1 series first heard Jun 2026 last heard 1 Jun

Ethan He’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 1 in all, peaking in Jun 2026 with 1.

newest first · ▶ plays the moment

Strong to weak.

It's kind of like strong to weak.

I guess from the modeling perspective,

The strong model, the teacher model is trying to model the image and videos of entire internet.

And that distribution is extremely complex.

But the step distill model is just trying to learn from the teacher.

The teacher is a model and the size is fixed.

The distribution is much simpler than the whole internet.

That's the intuition I have why step distillation can work.

So usually these models serve in productions.

They only run in a few steps.

In Cosmos, I believe we have four steps and eight steps.

If you do some simpler task, like image-to-image translation, it can even run in a few steps, like one step in Cosmos transfer.

Oh yeah, here it is.

It's already done.

So there are a few different approaches, for example, consistency model.

And there are also, actually, we shouldn't forget GAN.

So GAN, actually, that was an OG step dissolution because it trained just one step to begin with.

So actually a lot of, for example, there's a distribution matching dissolution which uses GAN as one of the laws for dissolution.

GAN just tells you, hey, generate an image, and then it has a discriminator to tell, is this image real or not?

Showing 281–300 of 722 · page 15 of 37