Variance-Adaptive Muon: Accelerating LLM Pretraining with NSR-Modulated and Variance-Scaled Momentum

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Jingru, Fan, Yibo, Li, Huan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!