Megadose AI progress, ranked and analyzed.

DeepLoop: Depth Scaling for Looped Transformers

· HF Daily Papers ·
Looped Transformers need residual scaling that accounts for repeated visits to the same parameters.

The paper formalizes how reused blocks change gradient behavior through a visit-alignment coefficient, `κR`. In the conservative aligned case, it says the scaling exponent should rise from `1/4` to `1/2` as loop count grows at fixed physical depth. DeepLoop keeps the Post-LN DeepNorm setup and sets `α=(2N)^{1/2}` and `β=(8N)^{-1/2}` for unrolled depth `N`. In GPT-style looped models at GPT-2 small and medium scale, it reports no penalty without revisits and better validation loss and downstream accuracy once recurrent depth is used. HF Daily Papers' note

score 4

Categories: Research