MMODELYST
People/Mingyi Hong
PPL

Mingyi Hong

Papers (3)
EMA-Nesterov: Stabilizing Nesterov's Lookahead for Accelerated Deep Learning OptimizationRethinking Muon Beyond Pretraining: Spectral Failures and High-Pass Remedies for VLA and RLVRRevisiting the Adam-SGD Gap in LLM Pre-Training: The Role of Large Effective Learning Rates