TY  - RPRT
TI  - Scale-Distribution Decoupling: Enabling Stable and Effective Training of Large Language Models
AU  - Ya Wang
AU  - Zhijian Zhuo
AU  - Yutao Zeng
AU  - Xun Zhou
AU  - Jian Yang
AU  - Xiaoqing Li
PY  - 2025
UR  - https://arxiv.org/abs/2502.15499
ID  - 2502.15499
ER  -