Frame selection is the whole game: notes on making LLMs watch video
leoaido.com"Dedup started as one comparator and grew a channel every time real footage embarrassed it."
Is it possible to ever read a piece by LLMs that isn't written by LLMs?
I'm so tired of opening another link and finding it's just Claude blogging under a pseudonym
I asked Claude to blog once (and only once); and it actually signed its named at the bottom in a beautiful act of self-attribution!
Was so proud; couldn't bring myself to do the violence of re-attribution: let it stand.
Oops, all Claudes!
perhaps a representation of movement rather than a series of frames would better model what people see when watching video - highly compressed formats might even be directly useful in this regard?
The actual project README appears more human-written and is actually pretty interesting: https://github.com/HUANGCHIHHUNGLeo/claude-real-video