LionMuon alternates Lion and Muon steps with shared dual-EMA buffer to Pareto-dominate existing optimizers in loss and compute on models up to 720M parameters.
Muon converges under heavy-tailed noise: Nonconvex h\"{o}lder-smooth empirical risk minimiza- tion.arXiv preprint arXiv:2603.15059
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LionMuon: Alternating Spectral and Sign Descent for Efficient Training
LionMuon alternates Lion and Muon steps with shared dual-EMA buffer to Pareto-dominate existing optimizers in loss and compute on models up to 720M parameters.