A Learning Rate Path Switching Training Paradigm for Version Updates of Large Language Models

A Learning Rate Path Switching Training Paradigm for Version Updates of Large Language Models Zhihao Wang author Shiyu Liu author Jianheng Huang author Wang Zheng author YiXuan Liao author Xiaoxin Chen author Junfeng Yao author Jinsong Su author 2024-11 text Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing Yaser Al-Onaizan editor Mohit Bansal editor Yun-Nung Chen editor Association for Computational Linguistics Miami, Florida, USA conference publication wang-etal-2024-learning-rate 10.18653/v1/2024.emnlp-main.752 https://aclanthology.org/2024.emnlp-main.752/ 2024-11 13581 13594