Can adding more loops to AI models deliver gains indefinitely?
A post introducing “Loop Scaling Laws” also asks what happens when looping meets mixture-of-experts.
TLDR
A post says frontier models like GPT-6 are reportedly using looping and points to “Loop Scaling Laws” for two questions: whether adding more loops can deliver gains indefinitely, and what happens when looping meets mixture-of-experts.
Combined views
84
1 Source, first seen ago