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
It can be characters or objects or even scenes.
Say like I want condition on Sean's selfie and the holding.
Holding a blade or whatever.
Yeah, you can put them there and the video models will generate the video and copy the context over.
So that can solve a lot of the problems there.
Like the long context problem, it doesn't need to have a very long context, but I feel like it's an intermediate solution.
It's cheating, yeah.
Yeah, the model should...
be able to selectively know where should I draw references.
So say if I want to generate a movie, I generate it autoregressive, like 10 seconds at a time or something.
And now this character appears, I can look back to where it first appeared and bring that back.
Yeah, this one.
I put the references.
Yeah, that's Optimus, Einstein, myself, Annie.
Yeah, I guess different labs have slightly different communication styles.
So I think it's like, this is like a mechanism to like select the context from the history and you might not put the entire history into the context.
For example, there's a paper called Frame Pack, which have a heuristic that the latest history, like the last one second, I put the entire history and
The history before that, I would compress it and make the video smaller.
So I follow this pattern, this buildable pattern that the maximum sequence length is fixed.
So the further you are from the current frame, you have a smaller image.
Showing 441–460 of 722 · page 23 of 37
← Previous
Next →