Muon Optimizer Underperforms AdamW in Controlled Matrix Factorization Study

Jul 16, 2026

A new preprint tests the Muon optimizer on low-rank matrix factorization, finding that it does not consistently outperform AdamW. The study suggests that Muon's previously reported advantages in large-scale deep learning may depend on factors such as scale, architecture, or hyperparameter sensitivity.

Why it matters: This work challenges assumptions about Muon's superiority and highlights the importance of controlled benchmarks for evaluating optimizers.

Full story at: arXiv Machine Learning ↗