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
So the prompt rewriter's task is to take
take user instruction and convert it to extremely detailed description of the video.
So because the video is a visual, the video distortion models, I would describe they're kind of dumb because they take the input instruction literally.
Because in the training process, remember that we have to describe the video as
as detailed as possible when we are creating the synthetic text pair.
So this model, they take those kind of instruction to generate the videos.
So when you're taking the user instructions, the user instruction is really very simple.
Just say a cat or something.
If you put a cat in
In the video model, they would take that instruction literally.
They would literally show a cat in maybe a white background because you didn't describe the background.
The cat is not moving because you didn't describe it.
It takes the instruction quite literally.
It's kind of dumb.
And the prompt for rider is actually a much bigger model, which is a language model that takes the...
the user instruction and expand it.
So the thinking process you mentioned is from there.
So if you look at like GPT image, like you generate a image in three minutes, three minutes is not all like a pixel generation.
A lot of time is spending in thinking.
So prompt rewriting now have evolved to
Showing 541–560 of 722 · page 28 of 37
← Previous
Next →