News for hackers. Stuff that matters.
Re:Attention at Constant Cost per Token via Symmetry-Aware T... by charleshmartin on Wednesday February 04, 2026 @09:14PM (#46894789)
Right. If the dynamics of training are governed by RG flow, then the best optimization path should remove redundant directions, as specified by the RG operator(s)

There are no comments on this comment yet.