Stepped MoE: reducing memory movement in large mixture-of-experts models
By arnav1993k · 2026-10-07 · 2 points · 1 comments
https://arxiv.org/abs/2610.07348
By arnav1993k · 2 points · 1 comments · on Hacker News, read on BetterNews.
Open the full discussion on BetterNews