CAME: Confidence-guided Adaptive Memory Efficient Optimization
Paper
•
2307.02047
•
Published
•
2
Optimizers
-
-
Practical Efficiency of Muon for Pretraining
Paper • 2505.02222 • Published • 40 -
AdaMuon: Adaptive Muon Optimizer
Paper • 2507.11005 • Published • 2 -
Muon is Scalable for LLM Training
Paper • 2502.16982 • Published • 10 -
Cautious Weight Decay
Paper • 2510.12402 • Published • 9 -
MSign: An Optimizer Preventing Training Instability in Large Language Models via Stable Rank Restoration
Paper • 2602.01734 • Published • 32