The winning argument is not about models but RLVR use cases rather than RLHF.
Huh, this might actually be a good lens for looking at why literary math got stuck at the philosophical level: some fixation on human feedback?
a blyg by Russell Antonie Pasetes
The winning argument is not about models but RLVR use cases rather than RLHF.
Huh, this might actually be a good lens for looking at why literary math got stuck at the philosophical level: some fixation on human feedback?