Back to papers
August 20, 2026cs.LGcs.AIcs.CL

Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts

Nayeon Kim, Hojin Lee, Yunju Bak, Jaesun Park, Boseop Kim
Read on arXivDownload PDF
HF Upvotes

36

Categories

cs.LG, cs.AI, cs.CL