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 monthsRecordings per month over the last 12 months — 1 in all, peaking in Jun 2026 with 1.
Appearances
I remember like a couple years ago, there's like six fingers or something.
Yeah, if you think about how these models were trained, like I mentioned before, GAN was in the training process.
The objective of GAN is the model generates an image.
And the model, there's a judge to tell if the image is real or not.
The model is trained to make the image more real.
So as the model becomes more and more advanced, it's going to be harder and harder.
For me personally, now I have to judge by if these videos have logical sense.
On that point, right?
Yeah, that's a good question.
Yeah, actually, I have a pretty big claim.
The visual intelligence are actually mostly coming from language.
Like these video models, especially from now, since the diffusion model technology is more mature, like every time you see there's some improvement on these models,
I would say mostly this, again, comes from language model, not coming from the video model itself, like the video distribution model themselves.
In Cosmos, that could be, typically these models, they have two parts.
Like there's a prompt rewriter or the prompt app sampler part.
I think in Cosmos,
In Cosmos, we use Lama or we use Mixerol.
And the Cosmos video model itself is only 7B.
And the language model is a prompt rewriter.
It's bigger than that.
Showing 521–540 of 722 · page 27 of 37
← Previous
Next →