@article{scalingandtransferabilityofannealingstra, title = {Scaling and Transferability of Annealing Strategies in Large Language Model Training}, author = {Siqi Wang and Zhengyu Chen and Teng Xiao and Zheqi Lv and Jinluan Yang and Xunliang Cai and Jingang Wang and Xiaomeng Li}, year = {2025}, eprint = {2512.13705}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2512.13705}, }