A Learning Rate Path Switching Training Paradigm for Version Updates of Large Language Models.
Zhihao Wang, Shiyu Liu, Jianheng Huang, Wang Zheng, Yixuan Liao, Xiaoxin Chen, Junfeng Yao, Jinsong Su
Browse the full EMNLP paper archive.
Zhihao Wang, Shiyu Liu, Jianheng Huang, Wang Zheng, Yixuan Liao, Xiaoxin Chen, Junfeng Yao, Jinsong Su
Browse the full EMNLP paper archive.